Instructions to use sandeeprdy1729/thrisul-parser with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use sandeeprdy1729/thrisul-parser with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("image-text-to-text", model="sandeeprdy1729/thrisul-parser", trust_remote_code=True)
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("sandeeprdy1729/thrisul-parser", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use sandeeprdy1729/thrisul-parser with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "sandeeprdy1729/thrisul-parser"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sandeeprdy1729/thrisul-parser",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker

docker model run hf.co/sandeeprdy1729/thrisul-parser

SGLang

How to use sandeeprdy1729/thrisul-parser with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "sandeeprdy1729/thrisul-parser" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sandeeprdy1729/thrisul-parser",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "sandeeprdy1729/thrisul-parser" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "sandeeprdy1729/thrisul-parser",
		"messages": [
			{
				"role": "user",
				"content": [
					{
						"type": "text",
						"text": "Describe this image in one sentence."
					},
					{
						"type": "image_url",
						"image_url": {
							"url": "https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg"
						}
					}
				]
			}
		]
	}'

Docker Model Runner
How to use sandeeprdy1729/thrisul-parser with Docker Model Runner:
```
docker model run hf.co/sandeeprdy1729/thrisul-parser
```

sandeeprdy1729 commited on 9 days ago

Commit

c92a0d2

verified ·

1 Parent(s): 8f9b1e9

benchmark artifacts (v1) (part 8)

Browse files

Files changed (38) hide show

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__atlantic.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__boxedtext.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__censored.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__colorprofile.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__crash.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__dash.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__docusigned.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__doubleslides.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__edit.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__edit2.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__gridofimages.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__ikea3.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__interface.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__json.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__justcode.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__lightblue.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__mangle.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__mark2.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__marks.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__missing.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__notATable.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__overlapping.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__redacted.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__redacted3.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__reverRo.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__skew3.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__stepbystep.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__templated.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__yellow.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__10k2col.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__14999.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__1815.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2cc.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2col.result.json +0 -0
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2col_fund_summary.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2col_paper.result.json +1 -1
benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2col_repport.result.json +1 -1
benchmarks/v1/results.json +275 -28

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__atlantic.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__boxedtext.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__boxedtext", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__boxedtext.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.\n\n| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.\n\n| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine."}, "output": {"task_type": "parse", "example_id": "text/text_misc__boxedtext", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.", "line_numbers": []}, {"page_index": 0, "markdown": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.\n\n| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.", "job_id": null}, "started_at": "2026-10-31T03:45:16.245909", "completed_at": "2026-10-31T03:45:16.245909", "latency_in_ms": 10939}

+ {"request": {"example_id": "text/text_misc__boxedtext", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__boxedtext.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.\n\n| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.\n\n| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine."}, "output": {"task_type": "parse", "example_id": "text/text_misc__boxedtext", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.", "line_numbers": []}, {"page_index": 0, "markdown": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.\n\n| Comments: La date de mise en ligne ainsi trouvee est 4 mois apres le milieu d'annee, ce qui correspond a la meme appreciation que celle effectuee par le chercheur meme si la mise en ligne s'est effectuee un mois plus tot que la date indiquee par le chercheur (31 octobre 2020 au lieu du 13 decembre). La date de mise en ligne ainsi obtenue semble plus coherente avec la structure du lien (qui comporte 2020/10)\n\n# Government Reviewer Opinion: Agree\n\nIBP Comment Les informations fournies par l'examinateur pair (notamment la m\u00e9thode d'ajout de la cha\u00eene \"INURL &as\\_qdr=y15\" dans la recherche Google) sont bien not\u00e9es.\n\n## MYR-4. If the MYR is published, what is the URL or weblink of the MYR?\n\n_Researchers should respond to this question if the document is published either within the time frame accepted by the OBS methodology or too late. If the document is not published at all, researchers should leave this question blank._\n\nAnswer: http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf<x_0.4678><y_0.3030>\n<x_0.0706><y_0.3664>Source: Revue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nComment: RAS\n\nPeer Reviewer Opinion: Agree\n\nMYR-5. If the MYR is published, are the numerical data contained in the MYR available in a machine readable format?\n\n_Material (data or content) is machine readable if it is in a format that can be easily processed by a computer, such as .csv, .xls/.xlsx, and .json. Numerical data found in PDFs, Word (.doc/.docx) and HTML files do not qualify as machine readable. See more at: http://opendatahandbook.org/glossary/en/terms/machine-readable/._\n\n_Option \"d\" applies if the MYR is not publicly available, therefore its machine readability cannot be assessed._\n\nAnswer: d. Not applicable\n\nRevue de milieu d\u2019ann\u00e9e 2020 http://www.dgfag.mg/wp-content/uploads/2020/10/RMA-2020.pdf\n\nLe fichier PDF n'est pas consid\u00e9r\u00e9 par l'\u00e9valuation comme format lisible\n\nPer Reviewer Opinion: Disagree Suggested Answer: c. No Comments: La reponse semble c plutot que d, puisque la RMA est bien publiee mais que les donnees numeriques ne sont pas disponibles dans un format lisible machine.", "job_id": null}, "started_at": "2026-10-31T12:47:34.700053", "completed_at": "2026-10-31T12:47:34.700053", "latency_in_ms": 10939}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__censored.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__censored", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__censored.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Document 3\n\n# Document 3", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Document 3\n\n# Document 3"}, "output": {"task_type": "parse", "example_id": "text/text_misc__censored", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Document 3", "line_numbers": []}, {"page_index": 0, "markdown": "# Document 3", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Document 3\n\n# Document 3", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:16.~~697567~~", "completed_at": "2026-10-~~31T03~~:45:16.~~697567~~", "latency_in_ms": 106083}

+ {"request": {"example_id": "text/text_misc__censored", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__censored.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Document 3\n\n# Document 3", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Document 3\n\n# Document 3"}, "output": {"task_type": "parse", "example_id": "text/text_misc__censored", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Document 3", "line_numbers": []}, {"page_index": 0, "markdown": "# Document 3", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Document 3\n\n# Document 3", "job_id": null}, "started_at": "2026-10-31T12:47:35.153261", "completed_at": "2026-10-31T12:47:35.153261", "latency_in_ms": 106083}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__colorprofile.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__crash.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__crash", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__crash.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |\n\n<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |\n\n<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |"}, "output": {"task_type": "parse", "example_id": "text/text_misc__crash", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |", "line_numbers": []}, {"page_index": 0, "markdown": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |\n\n<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:17.~~760621~~", "completed_at": "2026-10-~~31T03~~:45:17.~~760621~~", "latency_in_ms": 14695}

+ {"request": {"example_id": "text/text_misc__crash", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__crash.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |\n\n<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |\n\n<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |"}, "output": {"task_type": "parse", "example_id": "text/text_misc__crash", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |", "line_numbers": []}, {"page_index": 0, "markdown": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |\n\n<table><tr><th rowspan=\"2\"></th><th col| Category | East |\n|---|---|\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | OFFICER ID 10898 | AGENCY REPORT NUMBER I22052126 |\n| YEAR MONTH DAY 22 09 18 | HOUR MIN 14 18 | NCIC NO 0799 | GENCER ID 10898 | I22052126 |", "job_id": null}, "started_at": "2026-10-31T12:47:36.246953", "completed_at": "2026-10-31T12:47:36.246953", "latency_in_ms": 14695}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__dash.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__dash", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__dash.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "When estimating the risk adjustment transfer for the 2021 projection period, key variables in the risk\nadjustment transfer were estimated and applied against the Federal risk adjustment transfer formula.\nThese variables include the following:", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "When estimating the risk adjustment transfer for the 2021 projection period, key variables in the risk\nadjustment transfer were estimated and applied against the Federal risk adjustment transfer formula.\nThese variables include the following:"}, "output": {"task_type": "parse", "example_id": "text/text_misc__dash", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "When estimating the risk adjustment transfer for the 2021 projection period, key variables in the risk\nadjustment transfer were estimated and applied against the Federal risk adjustment transfer formula.\nThese variables include the following:", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "When estimating the risk adjustment transfer for the 2021 projection period, key variables in the risk\nadjustment transfer were estimated and applied against the Federal risk adjustment transfer formula.\nThese variables include the following:", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:18.~~089784~~", "completed_at": "2026-10-~~31T03~~:45:18.~~089784~~", "latency_in_ms": 106284}

+ {"request": {"example_id": "text/text_misc__dash", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__dash.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "When estimating the risk adjustment transfer for the 2021 projection period, key variables in the risk\nadjustment transfer were estimated and applied against the Federal risk adjustment transfer formula.\nThese variables include the following:", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "When estimating the risk adjustment transfer for the 2021 projection period, key variables in the risk\nadjustment transfer were estimated and applied against the Federal risk adjustment transfer formula.\nThese variables include the following:"}, "output": {"task_type": "parse", "example_id": "text/text_misc__dash", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "When estimating the risk adjustment transfer for the 2021 projection period, key variables in the risk\nadjustment transfer were estimated and applied against the Federal risk adjustment transfer formula.\nThese variables include the following:", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "When estimating the risk adjustment transfer for the 2021 projection period, key variables in the risk\nadjustment transfer were estimated and applied against the Federal risk adjustment transfer formula.\nThese variables include the following:", "job_id": null}, "started_at": "2026-10-31T12:47:36.573819", "completed_at": "2026-10-31T12:47:36.573819", "latency_in_ms": 106284}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__docusigned.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__doubleslides.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__edit.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__edit", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__edit.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>\n\n## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>\n\n## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>"}, "output": {"task_type": "parse", "example_id": "text/text_misc__edit", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>", "line_numbers": []}, {"page_index": 0, "markdown": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>\n\n## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>", "job_id": null}, "started_at": "2026-10-31T03:45:19.209543", "completed_at": "2026-10-31T03:45:19.209543", "latency_in_ms": 10593}

+ {"request": {"example_id": "text/text_misc__edit", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__edit.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>\n\n## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>\n\n## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>"}, "output": {"task_type": "parse", "example_id": "text/text_misc__edit", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>", "line_numbers": []}, {"page_index": 0, "markdown": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>\n\n## | Print out in text format the data collected from one or more meters, starting from a certain date and time, as per operator\u2019s instructions.<x_0.5123><y_0.0608|>| Providing access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.607|>| Access to SLDCs to provide user based access to view instantaneous MW data through WEB portal.<x_0.6061><y_0.618 |\n| The above software shall further ensure that absolutely no tampering (except erasing of complete data with password protection) of the collected metering data is possible during its handling by the PC.<x_0.6176><y_0.6300|>| The software shall be suitable for the commonly available PCs, (Windows) and shall be supplied to Employer in a compatible form to enable its easy loading into the PCs available (or to be installed by the Employer) at the various substations.<x_0.6176><y_0.6470|><class_Text>\n\n<x_0.0706><y_0.6578>| Provides access to RLDC for Energy Accounting software to download all meter related data from CDCS system in standard formats such as .csv, xlsx, text etc. through authenticated login on Web Portal. The CDCS system should provide standard API features (preferably RESTful APIs such JSON etc) to enable data exchange between IEMDC/RLDC and existing Energy Accounting software at RLDCs.<x_0.6176><y_0.6680|><class_Text>\n\n<x_0.0706><y_0.6770|>| Service of Work\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThis section provides detailed scope of work included in the bidder's scope, and facilities to be provided by station owner/employer.\n\nThe scope of work in complete conformity with subsequent sections of the specification shall include site survey, planning, design, engineering, manufacturing/integration, testing, supply, transportation & insurance, delivery at site, storage, installation, commissioning, demonstration for acceptance, training and documentation of AMR-MDP system including:<x_0.6067><y_0.7400|><class_Text>\n\n<x_0.0706><y_0.7530>a. Design Document for complete AMR -MDP System including software specifications.<x_0.6067><y_0.7670|><class_Text>\n\n<x_0.0706><y_0.7730|>d. Standard database (RDBMS) for storage of data at IEMDCs/RLDC.\n\ne. All cabling, wiring, terminations and interconnections of the equipment\n\n<class_Text>\n\n<x_0.0706><y_0.8420>h. Interconnection of main CDCS with Back-up CDCS on ICCP\n\ni. The complete system shall be in line with established cyber security standards.\n\n<class_Text>", "job_id": null}, "started_at": "2026-10-31T12:47:37.698839", "completed_at": "2026-10-31T12:47:37.698839", "latency_in_ms": 10593}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__edit2.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__gridofimages.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__gridofimages", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__gridofimages.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "| 053,20090718\u767d\u5c71\u796d\u308a042 | 054,20090718\u767d\u5c71\u796d\u308a043 | 055,20090718\u767d\u5c71\u796d\u308a044 | 056,20090718\u767d\u5c71\u796d\u308a045 | 057,20090718\u767d\u5c71\u796d\u308a046 |\n|---|---|---|---|\n| 058,20090718\u767d\u5c71\u796d\u308a047 | 059,20090718\u767d\u5c71\u796d\u308a048 | 059,20090718\u767d\u5c71\u796d\u308a049 | 060,20090718\u767d\u5c71\u796d\u308a050 | 061,20090718\u767d\u5c71\u796d\u308a051 |\n| 063,20090718\u767d\u5c71\u796d\u308a052 | 064,20090718\u767d\u5c71\u796d\u308a053 | 064,20090718\u767d\u5c71\u796d\u308a054 | 065,20090718\u767d\u5c71\u796d\u308a055 | 067,20090718\u767d\u5c71\u796d\u308a060 |\n| 068,20090718\u767d\u5c71\u796d\u308a057 | 069,20090718\u767d\u5c71\u796d\u308a058 | 069,20090718\u767d\u5c71\u796d\u308a059 | 070,20090718\u767d\u5c71\u796d\u308a060 | 072,20090718\u767d\u5c71\u796d\u308a061 |\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9773><table><tr><th rowspan=\"2\"></th><tr><th>054,20090718\u767d\u5c71\u796d\u308a043</th><td>055,20090718\u767d\u5c71\u796d\u308a044</td><td>056,20090718\u767d\u5c71\u796d\u308a045</td></tr><tr><td>058,20090718\u767d\u5c71\u796d\u308a047</td></tr><tr><td>059,20090718\u767d\u5c71\u796d\u308a048</td></tr><tr><td>060,20090718\u767d\u5c71\u796d\u308a050</td></tr><tr><td>061,20090718\u767d\u5c71\u796d\u308a051</td></tr><tr><td>063,20090718\u767d\u5c71\u796d\u308a052</td></tr><tr><td>064,20090718\u767d\u5c71\u796d\u308a053</td></tr><tr><td>065,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>066,20090718\u767d\u5c71\u796d\u308a054</td></tr><tr><td>067,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>073,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>074,20090718\u767d\u5c71\u796d\u308a063</td></tr><tr><td>075,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>076,20090718\u767d\u5c71\u796d\u308a065</td></tr><tr><td>077,20090718\u767d\u5c71\u796d\u308a066</td></tr><tr><td>078,20090718\u767d\u5c71\u796d\u308a067</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>080,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>081,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>082,20090718\u767d\u5c71\u796d\u308a071</td></tr><tr><td>083,20090718\u767d\u5c71\u796d\u308a072</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>085,20090718\u767d\u5c71\u796d\u308a074</td></tr><tr><td>086,20090718\u767d\u5c71\u796d\u308a075</td></tr><tr><td>087,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>088,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>089,20090718\u767d\u5c71\u796d\u308a078</td></tr><tr><td>090,20090718\u767d\u5c71\u796d\u308a079</td></tr><tr><td>091,20090718\u767d\u5c71\u796d\u308a080</td></tr><tr><td>092,20090718\u767d\u5c71\u796d\u308a081</td></tr><tr><td>093,20090718\u767d\u5c71\u796d\u308a082</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>095,20090718\u767d\u5c71\u796d\u308a084</td></tr><tr><td>096,20090718\u767d\u5c71\u796d\u308a085</td></tr><tr><td>097,20090718\u767d\u5c71\u796d\u308a086</td></tr></tr></table>\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9778><y_0.9970>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_0.9982><class_Text>\n\n<x_0.9706><y_0.9992>\u304a\u3051\u3089\u305b\u304f\u3060\u3055\u3044\u3002<x_0.9822><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.00988><class_Text>\n\n<x_0.9708><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1009>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1120>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1230><class_Text>\n\n<x_0.9708><y_0.9999>\u307e\u3059\u304b?<x_0.9722><y_1.1270>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1330><class_Text>\n\n<x_0.9708><y_1.1380>\u4e2d\u306e\u95a2\u4fc2\u304c\u3042\u308b\u3002<x_0.9738><y_1.1420>\u305d\u3046\u306a\u304b\u3082\u3042\u3063\u305f\u3002</we><y_1.1430><class_Text>\n\n<x_0.9708><y_1.1440>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002\n\n\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1650><class_Text>\n\n<x_0.9708><y_1.1670>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1770><class_Text>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "| 053,20090718\u767d\u5c71\u796d\u308a042 | 054,20090718\u767d\u5c71\u796d\u308a043 | 055,20090718\u767d\u5c71\u796d\u308a044 | 056,20090718\u767d\u5c71\u796d\u308a045 | 057,20090718\u767d\u5c71\u796d\u308a046 |\n|---|---|---|---|\n| 058,20090718\u767d\u5c71\u796d\u308a047 | 059,20090718\u767d\u5c71\u796d\u308a048 | 059,20090718\u767d\u5c71\u796d\u308a049 | 060,20090718\u767d\u5c71\u796d\u308a050 | 061,20090718\u767d\u5c71\u796d\u308a051 |\n| 063,20090718\u767d\u5c71\u796d\u308a052 | 064,20090718\u767d\u5c71\u796d\u308a053 | 064,20090718\u767d\u5c71\u796d\u308a054 | 065,20090718\u767d\u5c71\u796d\u308a055 | 067,20090718\u767d\u5c71\u796d\u308a060 |\n| 068,20090718\u767d\u5c71\u796d\u308a057 | 069,20090718\u767d\u5c71\u796d\u308a058 | 069,20090718\u767d\u5c71\u796d\u308a059 | 070,20090718\u767d\u5c71\u796d\u308a060 | 072,20090718\u767d\u5c71\u796d\u308a061 |\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9773><table><tr><th rowspan=\"2\"></th><tr><th>054,20090718\u767d\u5c71\u796d\u308a043</th><td>055,20090718\u767d\u5c71\u796d\u308a044</td><td>056,20090718\u767d\u5c71\u796d\u308a045</td></tr><tr><td>058,20090718\u767d\u5c71\u796d\u308a047</td></tr><tr><td>059,20090718\u767d\u5c71\u796d\u308a048</td></tr><tr><td>060,20090718\u767d\u5c71\u796d\u308a050</td></tr><tr><td>061,20090718\u767d\u5c71\u796d\u308a051</td></tr><tr><td>063,20090718\u767d\u5c71\u796d\u308a052</td></tr><tr><td>064,20090718\u767d\u5c71\u796d\u308a053</td></tr><tr><td>065,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>066,20090718\u767d\u5c71\u796d\u308a054</td></tr><tr><td>067,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>073,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>074,20090718\u767d\u5c71\u796d\u308a063</td></tr><tr><td>075,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>076,20090718\u767d\u5c71\u796d\u308a065</td></tr><tr><td>077,20090718\u767d\u5c71\u796d\u308a066</td></tr><tr><td>078,20090718\u767d\u5c71\u796d\u308a067</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>080,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>081,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>082,20090718\u767d\u5c71\u796d\u308a071</td></tr><tr><td>083,20090718\u767d\u5c71\u796d\u308a072</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>085,20090718\u767d\u5c71\u796d\u308a074</td></tr><tr><td>086,20090718\u767d\u5c71\u796d\u308a075</td></tr><tr><td>087,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>088,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>089,20090718\u767d\u5c71\u796d\u308a078</td></tr><tr><td>090,20090718\u767d\u5c71\u796d\u308a079</td></tr><tr><td>091,20090718\u767d\u5c71\u796d\u308a080</td></tr><tr><td>092,20090718\u767d\u5c71\u796d\u308a081</td></tr><tr><td>093,20090718\u767d\u5c71\u796d\u308a082</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>095,20090718\u767d\u5c71\u796d\u308a084</td></tr><tr><td>096,20090718\u767d\u5c71\u796d\u308a085</td></tr><tr><td>097,20090718\u767d\u5c71\u796d\u308a086</td></tr></tr></table>\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9778><y_0.9970>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_0.9982><class_Text>\n\n<x_0.9706><y_0.9992>\u304a\u3051\u3089\u305b\u304f\u3060\u3055\u3044\u3002<x_0.9822><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.00988><class_Text>\n\n<x_0.9708><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1009>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1120>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1230><class_Text>\n\n<x_0.9708><y_0.9999>\u307e\u3059\u304b?<x_0.9722><y_1.1270>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1330><class_Text>\n\n<x_0.9708><y_1.1380>\u4e2d\u306e\u95a2\u4fc2\u304c\u3042\u308b\u3002<x_0.9738><y_1.1420>\u305d\u3046\u306a\u304b\u3082\u3042\u3063\u305f\u3002</we><y_1.1430><class_Text>\n\n<x_0.9708><y_1.1440>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002\n\n\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1650><class_Text>\n\n<x_0.9708><y_1.1670>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1770><class_Text>"}, "output": {"task_type": "parse", "example_id": "text/text_misc__gridofimages", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "| 053,20090718\u767d\u5c71\u796d\u308a042 | 054,20090718\u767d\u5c71\u796d\u308a043 | 055,20090718\u767d\u5c71\u796d\u308a044 | 056,20090718\u767d\u5c71\u796d\u308a045 | 057,20090718\u767d\u5c71\u796d\u308a046 |\n|---|---|---|---|\n| 058,20090718\u767d\u5c71\u796d\u308a047 | 059,20090718\u767d\u5c71\u796d\u308a048 | 059,20090718\u767d\u5c71\u796d\u308a049 | 060,20090718\u767d\u5c71\u796d\u308a050 | 061,20090718\u767d\u5c71\u796d\u308a051 |\n| 063,20090718\u767d\u5c71\u796d\u308a052 | 064,20090718\u767d\u5c71\u796d\u308a053 | 064,20090718\u767d\u5c71\u796d\u308a054 | 065,20090718\u767d\u5c71\u796d\u308a055 | 067,20090718\u767d\u5c71\u796d\u308a060 |\n| 068,20090718\u767d\u5c71\u796d\u308a057 | 069,20090718\u767d\u5c71\u796d\u308a058 | 069,20090718\u767d\u5c71\u796d\u308a059 | 070,20090718\u767d\u5c71\u796d\u308a060 | 072,20090718\u767d\u5c71\u796d\u308a061 |\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9773><table><tr><th rowspan=\"2\"></th><tr><th>054,20090718\u767d\u5c71\u796d\u308a043</th><td>055,20090718\u767d\u5c71\u796d\u308a044</td><td>056,20090718\u767d\u5c71\u796d\u308a045</td></tr><tr><td>058,20090718\u767d\u5c71\u796d\u308a047</td></tr><tr><td>059,20090718\u767d\u5c71\u796d\u308a048</td></tr><tr><td>060,20090718\u767d\u5c71\u796d\u308a050</td></tr><tr><td>061,20090718\u767d\u5c71\u796d\u308a051</td></tr><tr><td>063,20090718\u767d\u5c71\u796d\u308a052</td></tr><tr><td>064,20090718\u767d\u5c71\u796d\u308a053</td></tr><tr><td>065,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>066,20090718\u767d\u5c71\u796d\u308a054</td></tr><tr><td>067,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>073,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>074,20090718\u767d\u5c71\u796d\u308a063</td></tr><tr><td>075,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>076,20090718\u767d\u5c71\u796d\u308a065</td></tr><tr><td>077,20090718\u767d\u5c71\u796d\u308a066</td></tr><tr><td>078,20090718\u767d\u5c71\u796d\u308a067</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>080,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>081,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>082,20090718\u767d\u5c71\u796d\u308a071</td></tr><tr><td>083,20090718\u767d\u5c71\u796d\u308a072</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>085,20090718\u767d\u5c71\u796d\u308a074</td></tr><tr><td>086,20090718\u767d\u5c71\u796d\u308a075</td></tr><tr><td>087,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>088,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>089,20090718\u767d\u5c71\u796d\u308a078</td></tr><tr><td>090,20090718\u767d\u5c71\u796d\u308a079</td></tr><tr><td>091,20090718\u767d\u5c71\u796d\u308a080</td></tr><tr><td>092,20090718\u767d\u5c71\u796d\u308a081</td></tr><tr><td>093,20090718\u767d\u5c71\u796d\u308a082</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>095,20090718\u767d\u5c71\u796d\u308a084</td></tr><tr><td>096,20090718\u767d\u5c71\u796d\u308a085</td></tr><tr><td>097,20090718\u767d\u5c71\u796d\u308a086</td></tr></tr></table>\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9778><y_0.9970>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_0.9982><class_Text>\n\n<x_0.9706><y_0.9992>\u304a\u3051\u3089\u305b\u304f\u3060\u3055\u3044\u3002<x_0.9822><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.00988><class_Text>\n\n<x_0.9708><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1009>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1120>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1230><class_Text>\n\n<x_0.9708><y_0.9999>\u307e\u3059\u304b?<x_0.9722><y_1.1270>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1330><class_Text>\n\n<x_0.9708><y_1.1380>\u4e2d\u306e\u95a2\u4fc2\u304c\u3042\u308b\u3002<x_0.9738><y_1.1420>\u305d\u3046\u306a\u304b\u3082\u3042\u3063\u305f\u3002</we><y_1.1430><class_Text>\n\n<x_0.9708><y_1.1440>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002\n\n\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1650><class_Text>\n\n<x_0.9708><y_1.1670>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1770><class_Text>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "| 053,20090718\u767d\u5c71\u796d\u308a042 | 054,20090718\u767d\u5c71\u796d\u308a043 | 055,20090718\u767d\u5c71\u796d\u308a044 | 056,20090718\u767d\u5c71\u796d\u308a045 | 057,20090718\u767d\u5c71\u796d\u308a046 |\n|---|---|---|---|\n| 058,20090718\u767d\u5c71\u796d\u308a047 | 059,20090718\u767d\u5c71\u796d\u308a048 | 059,20090718\u767d\u5c71\u796d\u308a049 | 060,20090718\u767d\u5c71\u796d\u308a050 | 061,20090718\u767d\u5c71\u796d\u308a051 |\n| 063,20090718\u767d\u5c71\u796d\u308a052 | 064,20090718\u767d\u5c71\u796d\u308a053 | 064,20090718\u767d\u5c71\u796d\u308a054 | 065,20090718\u767d\u5c71\u796d\u308a055 | 067,20090718\u767d\u5c71\u796d\u308a060 |\n| 068,20090718\u767d\u5c71\u796d\u308a057 | 069,20090718\u767d\u5c71\u796d\u308a058 | 069,20090718\u767d\u5c71\u796d\u308a059 | 070,20090718\u767d\u5c71\u796d\u308a060 | 072,20090718\u767d\u5c71\u796d\u308a061 |\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9773><table><tr><th rowspan=\"2\"></th><tr><th>054,20090718\u767d\u5c71\u796d\u308a043</th><td>055,20090718\u767d\u5c71\u796d\u308a044</td><td>056,20090718\u767d\u5c71\u796d\u308a045</td></tr><tr><td>058,20090718\u767d\u5c71\u796d\u308a047</td></tr><tr><td>059,20090718\u767d\u5c71\u796d\u308a048</td></tr><tr><td>060,20090718\u767d\u5c71\u796d\u308a050</td></tr><tr><td>061,20090718\u767d\u5c71\u796d\u308a051</td></tr><tr><td>063,20090718\u767d\u5c71\u796d\u308a052</td></tr><tr><td>064,20090718\u767d\u5c71\u796d\u308a053</td></tr><tr><td>065,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>066,20090718\u767d\u5c71\u796d\u308a054</td></tr><tr><td>067,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>073,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>074,20090718\u767d\u5c71\u796d\u308a063</td></tr><tr><td>075,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>076,20090718\u767d\u5c71\u796d\u308a065</td></tr><tr><td>077,20090718\u767d\u5c71\u796d\u308a066</td></tr><tr><td>078,20090718\u767d\u5c71\u796d\u308a067</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>080,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>081,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>082,20090718\u767d\u5c71\u796d\u308a071</td></tr><tr><td>083,20090718\u767d\u5c71\u796d\u308a072</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>085,20090718\u767d\u5c71\u796d\u308a074</td></tr><tr><td>086,20090718\u767d\u5c71\u796d\u308a075</td></tr><tr><td>087,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>088,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>089,20090718\u767d\u5c71\u796d\u308a078</td></tr><tr><td>090,20090718\u767d\u5c71\u796d\u308a079</td></tr><tr><td>091,20090718\u767d\u5c71\u796d\u308a080</td></tr><tr><td>092,20090718\u767d\u5c71\u796d\u308a081</td></tr><tr><td>093,20090718\u767d\u5c71\u796d\u308a082</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>095,20090718\u767d\u5c71\u796d\u308a084</td></tr><tr><td>096,20090718\u767d\u5c71\u796d\u308a085</td></tr><tr><td>097,20090718\u767d\u5c71\u796d\u308a086</td></tr></tr></table>\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9778><y_0.9970>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_0.9982><class_Text>\n\n<x_0.9706><y_0.9992>\u304a\u3051\u3089\u305b\u304f\u3060\u3055\u3044\u3002<x_0.9822><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.00988><class_Text>\n\n<x_0.9708><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1009>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1120>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1230><class_Text>\n\n<x_0.9708><y_0.9999>\u307e\u3059\u304b?<x_0.9722><y_1.1270>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1330><class_Text>\n\n<x_0.9708><y_1.1380>\u4e2d\u306e\u95a2\u4fc2\u304c\u3042\u308b\u3002<x_0.9738><y_1.1420>\u305d\u3046\u306a\u304b\u3082\u3042\u3063\u305f\u3002</we><y_1.1430><class_Text>\n\n<x_0.9708><y_1.1440>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002\n\n\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1650><class_Text>\n\n<x_0.9708><y_1.1670>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1770><class_Text>", "job_id": null}, "started_at": "2026-10-31T03:45:19.935163", "completed_at": "2026-10-31T03:45:19.935163", "latency_in_ms": 20498}

+ {"request": {"example_id": "text/text_misc__gridofimages", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__gridofimages.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "| 053,20090718\u767d\u5c71\u796d\u308a042 | 054,20090718\u767d\u5c71\u796d\u308a043 | 055,20090718\u767d\u5c71\u796d\u308a044 | 056,20090718\u767d\u5c71\u796d\u308a045 | 057,20090718\u767d\u5c71\u796d\u308a046 |\n|---|---|---|---|\n| 058,20090718\u767d\u5c71\u796d\u308a047 | 059,20090718\u767d\u5c71\u796d\u308a048 | 059,20090718\u767d\u5c71\u796d\u308a049 | 060,20090718\u767d\u5c71\u796d\u308a050 | 061,20090718\u767d\u5c71\u796d\u308a051 |\n| 063,20090718\u767d\u5c71\u796d\u308a052 | 064,20090718\u767d\u5c71\u796d\u308a053 | 064,20090718\u767d\u5c71\u796d\u308a054 | 065,20090718\u767d\u5c71\u796d\u308a055 | 067,20090718\u767d\u5c71\u796d\u308a060 |\n| 068,20090718\u767d\u5c71\u796d\u308a057 | 069,20090718\u767d\u5c71\u796d\u308a058 | 069,20090718\u767d\u5c71\u796d\u308a059 | 070,20090718\u767d\u5c71\u796d\u308a060 | 072,20090718\u767d\u5c71\u796d\u308a061 |\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9773><table><tr><th rowspan=\"2\"></th><tr><th>054,20090718\u767d\u5c71\u796d\u308a043</th><td>055,20090718\u767d\u5c71\u796d\u308a044</td><td>056,20090718\u767d\u5c71\u796d\u308a045</td></tr><tr><td>058,20090718\u767d\u5c71\u796d\u308a047</td></tr><tr><td>059,20090718\u767d\u5c71\u796d\u308a048</td></tr><tr><td>060,20090718\u767d\u5c71\u796d\u308a050</td></tr><tr><td>061,20090718\u767d\u5c71\u796d\u308a051</td></tr><tr><td>063,20090718\u767d\u5c71\u796d\u308a052</td></tr><tr><td>064,20090718\u767d\u5c71\u796d\u308a053</td></tr><tr><td>065,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>066,20090718\u767d\u5c71\u796d\u308a054</td></tr><tr><td>067,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>073,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>074,20090718\u767d\u5c71\u796d\u308a063</td></tr><tr><td>075,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>076,20090718\u767d\u5c71\u796d\u308a065</td></tr><tr><td>077,20090718\u767d\u5c71\u796d\u308a066</td></tr><tr><td>078,20090718\u767d\u5c71\u796d\u308a067</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>080,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>081,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>082,20090718\u767d\u5c71\u796d\u308a071</td></tr><tr><td>083,20090718\u767d\u5c71\u796d\u308a072</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>085,20090718\u767d\u5c71\u796d\u308a074</td></tr><tr><td>086,20090718\u767d\u5c71\u796d\u308a075</td></tr><tr><td>087,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>088,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>089,20090718\u767d\u5c71\u796d\u308a078</td></tr><tr><td>090,20090718\u767d\u5c71\u796d\u308a079</td></tr><tr><td>091,20090718\u767d\u5c71\u796d\u308a080</td></tr><tr><td>092,20090718\u767d\u5c71\u796d\u308a081</td></tr><tr><td>093,20090718\u767d\u5c71\u796d\u308a082</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>095,20090718\u767d\u5c71\u796d\u308a084</td></tr><tr><td>096,20090718\u767d\u5c71\u796d\u308a085</td></tr><tr><td>097,20090718\u767d\u5c71\u796d\u308a086</td></tr></tr></table>\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9778><y_0.9970>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_0.9982><class_Text>\n\n<x_0.9706><y_0.9992>\u304a\u3051\u3089\u305b\u304f\u3060\u3055\u3044\u3002<x_0.9822><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.00988><class_Text>\n\n<x_0.9708><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1009>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1120>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1230><class_Text>\n\n<x_0.9708><y_0.9999>\u307e\u3059\u304b?<x_0.9722><y_1.1270>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1330><class_Text>\n\n<x_0.9708><y_1.1380>\u4e2d\u306e\u95a2\u4fc2\u304c\u3042\u308b\u3002<x_0.9738><y_1.1420>\u305d\u3046\u306a\u304b\u3082\u3042\u3063\u305f\u3002</we><y_1.1430><class_Text>\n\n<x_0.9708><y_1.1440>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002\n\n\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1650><class_Text>\n\n<x_0.9708><y_1.1670>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1770><class_Text>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "| 053,20090718\u767d\u5c71\u796d\u308a042 | 054,20090718\u767d\u5c71\u796d\u308a043 | 055,20090718\u767d\u5c71\u796d\u308a044 | 056,20090718\u767d\u5c71\u796d\u308a045 | 057,20090718\u767d\u5c71\u796d\u308a046 |\n|---|---|---|---|\n| 058,20090718\u767d\u5c71\u796d\u308a047 | 059,20090718\u767d\u5c71\u796d\u308a048 | 059,20090718\u767d\u5c71\u796d\u308a049 | 060,20090718\u767d\u5c71\u796d\u308a050 | 061,20090718\u767d\u5c71\u796d\u308a051 |\n| 063,20090718\u767d\u5c71\u796d\u308a052 | 064,20090718\u767d\u5c71\u796d\u308a053 | 064,20090718\u767d\u5c71\u796d\u308a054 | 065,20090718\u767d\u5c71\u796d\u308a055 | 067,20090718\u767d\u5c71\u796d\u308a060 |\n| 068,20090718\u767d\u5c71\u796d\u308a057 | 069,20090718\u767d\u5c71\u796d\u308a058 | 069,20090718\u767d\u5c71\u796d\u308a059 | 070,20090718\u767d\u5c71\u796d\u308a060 | 072,20090718\u767d\u5c71\u796d\u308a061 |\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9773><table><tr><th rowspan=\"2\"></th><tr><th>054,20090718\u767d\u5c71\u796d\u308a043</th><td>055,20090718\u767d\u5c71\u796d\u308a044</td><td>056,20090718\u767d\u5c71\u796d\u308a045</td></tr><tr><td>058,20090718\u767d\u5c71\u796d\u308a047</td></tr><tr><td>059,20090718\u767d\u5c71\u796d\u308a048</td></tr><tr><td>060,20090718\u767d\u5c71\u796d\u308a050</td></tr><tr><td>061,20090718\u767d\u5c71\u796d\u308a051</td></tr><tr><td>063,20090718\u767d\u5c71\u796d\u308a052</td></tr><tr><td>064,20090718\u767d\u5c71\u796d\u308a053</td></tr><tr><td>065,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>066,20090718\u767d\u5c71\u796d\u308a054</td></tr><tr><td>067,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>073,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>074,20090718\u767d\u5c71\u796d\u308a063</td></tr><tr><td>075,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>076,20090718\u767d\u5c71\u796d\u308a065</td></tr><tr><td>077,20090718\u767d\u5c71\u796d\u308a066</td></tr><tr><td>078,20090718\u767d\u5c71\u796d\u308a067</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>080,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>081,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>082,20090718\u767d\u5c71\u796d\u308a071</td></tr><tr><td>083,20090718\u767d\u5c71\u796d\u308a072</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>085,20090718\u767d\u5c71\u796d\u308a074</td></tr><tr><td>086,20090718\u767d\u5c71\u796d\u308a075</td></tr><tr><td>087,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>088,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>089,20090718\u767d\u5c71\u796d\u308a078</td></tr><tr><td>090,20090718\u767d\u5c71\u796d\u308a079</td></tr><tr><td>091,20090718\u767d\u5c71\u796d\u308a080</td></tr><tr><td>092,20090718\u767d\u5c71\u796d\u308a081</td></tr><tr><td>093,20090718\u767d\u5c71\u796d\u308a082</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>095,20090718\u767d\u5c71\u796d\u308a084</td></tr><tr><td>096,20090718\u767d\u5c71\u796d\u308a085</td></tr><tr><td>097,20090718\u767d\u5c71\u796d\u308a086</td></tr></tr></table>\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9778><y_0.9970>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_0.9982><class_Text>\n\n<x_0.9706><y_0.9992>\u304a\u3051\u3089\u305b\u304f\u3060\u3055\u3044\u3002<x_0.9822><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.00988><class_Text>\n\n<x_0.9708><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1009>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1120>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1230><class_Text>\n\n<x_0.9708><y_0.9999>\u307e\u3059\u304b?<x_0.9722><y_1.1270>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1330><class_Text>\n\n<x_0.9708><y_1.1380>\u4e2d\u306e\u95a2\u4fc2\u304c\u3042\u308b\u3002<x_0.9738><y_1.1420>\u305d\u3046\u306a\u304b\u3082\u3042\u3063\u305f\u3002</we><y_1.1430><class_Text>\n\n<x_0.9708><y_1.1440>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002\n\n\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1650><class_Text>\n\n<x_0.9708><y_1.1670>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1770><class_Text>"}, "output": {"task_type": "parse", "example_id": "text/text_misc__gridofimages", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "| 053,20090718\u767d\u5c71\u796d\u308a042 | 054,20090718\u767d\u5c71\u796d\u308a043 | 055,20090718\u767d\u5c71\u796d\u308a044 | 056,20090718\u767d\u5c71\u796d\u308a045 | 057,20090718\u767d\u5c71\u796d\u308a046 |\n|---|---|---|---|\n| 058,20090718\u767d\u5c71\u796d\u308a047 | 059,20090718\u767d\u5c71\u796d\u308a048 | 059,20090718\u767d\u5c71\u796d\u308a049 | 060,20090718\u767d\u5c71\u796d\u308a050 | 061,20090718\u767d\u5c71\u796d\u308a051 |\n| 063,20090718\u767d\u5c71\u796d\u308a052 | 064,20090718\u767d\u5c71\u796d\u308a053 | 064,20090718\u767d\u5c71\u796d\u308a054 | 065,20090718\u767d\u5c71\u796d\u308a055 | 067,20090718\u767d\u5c71\u796d\u308a060 |\n| 068,20090718\u767d\u5c71\u796d\u308a057 | 069,20090718\u767d\u5c71\u796d\u308a058 | 069,20090718\u767d\u5c71\u796d\u308a059 | 070,20090718\u767d\u5c71\u796d\u308a060 | 072,20090718\u767d\u5c71\u796d\u308a061 |\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9773><table><tr><th rowspan=\"2\"></th><tr><th>054,20090718\u767d\u5c71\u796d\u308a043</th><td>055,20090718\u767d\u5c71\u796d\u308a044</td><td>056,20090718\u767d\u5c71\u796d\u308a045</td></tr><tr><td>058,20090718\u767d\u5c71\u796d\u308a047</td></tr><tr><td>059,20090718\u767d\u5c71\u796d\u308a048</td></tr><tr><td>060,20090718\u767d\u5c71\u796d\u308a050</td></tr><tr><td>061,20090718\u767d\u5c71\u796d\u308a051</td></tr><tr><td>063,20090718\u767d\u5c71\u796d\u308a052</td></tr><tr><td>064,20090718\u767d\u5c71\u796d\u308a053</td></tr><tr><td>065,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>066,20090718\u767d\u5c71\u796d\u308a054</td></tr><tr><td>067,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>073,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>074,20090718\u767d\u5c71\u796d\u308a063</td></tr><tr><td>075,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>076,20090718\u767d\u5c71\u796d\u308a065</td></tr><tr><td>077,20090718\u767d\u5c71\u796d\u308a066</td></tr><tr><td>078,20090718\u767d\u5c71\u796d\u308a067</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>080,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>081,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>082,20090718\u767d\u5c71\u796d\u308a071</td></tr><tr><td>083,20090718\u767d\u5c71\u796d\u308a072</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>085,20090718\u767d\u5c71\u796d\u308a074</td></tr><tr><td>086,20090718\u767d\u5c71\u796d\u308a075</td></tr><tr><td>087,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>088,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>089,20090718\u767d\u5c71\u796d\u308a078</td></tr><tr><td>090,20090718\u767d\u5c71\u796d\u308a079</td></tr><tr><td>091,20090718\u767d\u5c71\u796d\u308a080</td></tr><tr><td>092,20090718\u767d\u5c71\u796d\u308a081</td></tr><tr><td>093,20090718\u767d\u5c71\u796d\u308a082</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>095,20090718\u767d\u5c71\u796d\u308a084</td></tr><tr><td>096,20090718\u767d\u5c71\u796d\u308a085</td></tr><tr><td>097,20090718\u767d\u5c71\u796d\u308a086</td></tr></tr></table>\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9778><y_0.9970>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_0.9982><class_Text>\n\n<x_0.9706><y_0.9992>\u304a\u3051\u3089\u305b\u304f\u3060\u3055\u3044\u3002<x_0.9822><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.00988><class_Text>\n\n<x_0.9708><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1009>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1120>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1230><class_Text>\n\n<x_0.9708><y_0.9999>\u307e\u3059\u304b?<x_0.9722><y_1.1270>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1330><class_Text>\n\n<x_0.9708><y_1.1380>\u4e2d\u306e\u95a2\u4fc2\u304c\u3042\u308b\u3002<x_0.9738><y_1.1420>\u305d\u3046\u306a\u304b\u3082\u3042\u3063\u305f\u3002</we><y_1.1430><class_Text>\n\n<x_0.9708><y_1.1440>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002\n\n\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1650><class_Text>\n\n<x_0.9708><y_1.1670>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1770><class_Text>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "| 053,20090718\u767d\u5c71\u796d\u308a042 | 054,20090718\u767d\u5c71\u796d\u308a043 | 055,20090718\u767d\u5c71\u796d\u308a044 | 056,20090718\u767d\u5c71\u796d\u308a045 | 057,20090718\u767d\u5c71\u796d\u308a046 |\n|---|---|---|---|\n| 058,20090718\u767d\u5c71\u796d\u308a047 | 059,20090718\u767d\u5c71\u796d\u308a048 | 059,20090718\u767d\u5c71\u796d\u308a049 | 060,20090718\u767d\u5c71\u796d\u308a050 | 061,20090718\u767d\u5c71\u796d\u308a051 |\n| 063,20090718\u767d\u5c71\u796d\u308a052 | 064,20090718\u767d\u5c71\u796d\u308a053 | 064,20090718\u767d\u5c71\u796d\u308a054 | 065,20090718\u767d\u5c71\u796d\u308a055 | 067,20090718\u767d\u5c71\u796d\u308a060 |\n| 068,20090718\u767d\u5c71\u796d\u308a057 | 069,20090718\u767d\u5c71\u796d\u308a058 | 069,20090718\u767d\u5c71\u796d\u308a059 | 070,20090718\u767d\u5c71\u796d\u308a060 | 072,20090718\u767d\u5c71\u796d\u308a061 |\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9773><table><tr><th rowspan=\"2\"></th><tr><th>054,20090718\u767d\u5c71\u796d\u308a043</th><td>055,20090718\u767d\u5c71\u796d\u308a044</td><td>056,20090718\u767d\u5c71\u796d\u308a045</td></tr><tr><td>058,20090718\u767d\u5c71\u796d\u308a047</td></tr><tr><td>059,20090718\u767d\u5c71\u796d\u308a048</td></tr><tr><td>060,20090718\u767d\u5c71\u796d\u308a050</td></tr><tr><td>061,20090718\u767d\u5c71\u796d\u308a051</td></tr><tr><td>063,20090718\u767d\u5c71\u796d\u308a052</td></tr><tr><td>064,20090718\u767d\u5c71\u796d\u308a053</td></tr><tr><td>065,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>066,20090718\u767d\u5c71\u796d\u308a054</td></tr><tr><td>067,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>073,20090718\u767d\u5c71\u796d\u308a062</td></tr><tr><td>074,20090718\u767d\u5c71\u796d\u308a063</td></tr><tr><td>075,20090718\u767d\u5c71\u796d\u308a064</td></tr><tr><td>076,20090718\u767d\u5c71\u796d\u308a065</td></tr><tr><td>077,20090718\u767d\u5c71\u796d\u308a066</td></tr><tr><td>078,20090718\u767d\u5c71\u796d\u308a067</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>079,20090718\u767d\u5c71\u796d\u308a068</td></tr><tr><td>080,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>081,20090718\u767d\u5c71\u796d\u308a070</td></tr><tr><td>082,20090718\u767d\u5c71\u796d\u308a071</td></tr><tr><td>083,20090718\u767d\u5c71\u796d\u308a072</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>084,20090718\u767d\u5c71\u796d\u308a073</td></tr><tr><td>085,20090718\u767d\u5c71\u796d\u308a074</td></tr><tr><td>086,20090718\u767d\u5c71\u796d\u308a075</td></tr><tr><td>087,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>088,20090718\u767d\u5c71\u796d\u308a077</td></tr><tr><td>089,20090718\u767d\u5c71\u796d\u308a078</td></tr><tr><td>090,20090718\u767d\u5c71\u796d\u308a079</td></tr><tr><td>091,20090718\u767d\u5c71\u796d\u308a080</td></tr><tr><td>092,20090718\u767d\u5c71\u796d\u308a081</td></tr><tr><td>093,20090718\u767d\u5c71\u796d\u308a082</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>094,20090718\u767d\u5c71\u796d\u308a083</td></tr><tr><td>095,20090718\u767d\u5c71\u796d\u308a084</td></tr><tr><td>096,20090718\u767d\u5c71\u796d\u308a085</td></tr><tr><td>097,20090718\u767d\u5c71\u796d\u308a086</td></tr></tr></table>\n\n\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<x_0.9778><y_0.9970>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_0.9982><class_Text>\n\n<x_0.9706><y_0.9992>\u304a\u3051\u3089\u305b\u304f\u3060\u3055\u3044\u3002<x_0.9822><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.00988><class_Text>\n\n<x_0.9708><y_0.9998>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1009>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1120>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1230><class_Text>\n\n<x_0.9708><y_0.9999>\u307e\u3059\u304b?<x_0.9722><y_1.1270>\u305d\u3046\u306a\u304b\u3067\u3042\u3044\u3002<y_1.1330><class_Text>\n\n<x_0.9708><y_1.1380>\u4e2d\u306e\u95a2\u4fc2\u304c\u3042\u308b\u3002<x_0.9738><y_1.1420>\u305d\u3046\u306a\u304b\u3082\u3042\u3063\u305f\u3002</we><y_1.1430><class_Text>\n\n<x_0.9708><y_1.1440>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002\n\n\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1650><class_Text>\n\n<x_0.9708><y_1.1670>\u4e2d\u306b\u95a2\u308f\u306a\u3044\u3002\u3053\u3068\u306f\u306a\u3044\u3067\u3059\u3002<y_1.1770><class_Text>", "job_id": null}, "started_at": "2026-10-31T12:47:38.430647", "completed_at": "2026-10-31T12:47:38.430647", "latency_in_ms": 20498}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__ikea3.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__interface.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__interface", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__interface.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com\n\n# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com\n\n# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com"}, "output": {"task_type": "parse", "example_id": "text/text_misc__interface", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com", "line_numbers": []}, {"page_index": 0, "markdown": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com\n\n# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:20.~~858686~~", "completed_at": "2026-10-~~31T03~~:45:20.~~858686~~", "latency_in_ms": 5042}

+ {"request": {"example_id": "text/text_misc__interface", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__interface.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com\n\n# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com\n\n# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com"}, "output": {"task_type": "parse", "example_id": "text/text_misc__interface", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com", "line_numbers": []}, {"page_index": 0, "markdown": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com\n\n# GovCreateMpcrrDocument Page 10 of 28\n\nPage 10 of 28\n\n## | Operated | Demo Builder | Enter |\n|---|---|---|\n| User | Government | Depormentation | Linkap |\n| Logs | Exit | Ext |\n| Acceptor - Micro-Purchase Receiving Report | Contract | Document Flow Info | Routing | Data Capture |\n| Work For Rep | Work For Secondary | Tax | Counts |\n| Mark For Rep | Mark For Secondary | Mark For Secondary | Mark For Secondary |\n| The Mark For Page is displayed. The Acceptor can use the Mark For tab to enter comments for the Mark For Rep and Mark For Secondary. | Submit | Save Drift Document | NIS |<x_0.7476><y_0.5776><y_0.6496><class_Section-header>\n\n<x_0.0706><y_0.6521><table><y_0.6655><table><table><y_0.6727><class_Table>\n\n<x_0.0706><y_0.6755>scene 20\n\nThe Acceptor can navigate to the Comments page by clicking the Comments tab.\n\nBuilt with Tanida Demo Builder www.demo-builder.com", "job_id": null}, "started_at": "2026-10-31T12:47:39.364189", "completed_at": "2026-10-31T12:47:39.364189", "latency_in_ms": 5042}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__json.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__justcode.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__justcode", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__justcode.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>\n\n# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>\n\n# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>"}, "output": {"task_type": "parse", "example_id": "text/text_misc__justcode", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>", "line_numbers": []}, {"page_index": 0, "markdown": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>\n\n# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>", "job_id": null}, "started_at": "2026-10-31T03:45:21.572015", "completed_at": "2026-10-31T03:45:21.572015", "latency_in_ms": 25781}

+ {"request": {"example_id": "text/text_misc__justcode", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__justcode.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>\n\n# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>\n\n# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>"}, "output": {"task_type": "parse", "example_id": "text/text_misc__justcode", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>", "line_numbers": []}, {"page_index": 0, "markdown": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>\n\n# list of access JPEGs<x_0.3692><y_0.0661>self.jpg_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.jpg')]) self.jpg_total = len(self.jpg_list)<x_0.7676><y_0.0738> #List of access PDFs<x_0.3482><y_0.0938>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of expected files<x_0.3675><y_0.1061>self.expected_files_list = self.pm_list + self.mm_list + self.jpg_list + self.pdf_list self.expected_files_list]= len(self.expected_files_list)<x_0.7988><y_0.2124><table><tr><th># List of access JPEGs</th></tr><tr><td>self.pdf_list = sorted([x **for** x **in** self.files_list if str(x).endswith('_ac.pdf')]) self.pdf_total = len(self.pdf_list)\n\n# list of extra files<x_0.3675><y_0.2258>self.extra_files_list = sorted([x **for** x **in** self.files_list if x **not in** self.expected_files_list]) self.extra_files_list total = len(self.extra_files_list)\n\nIn [ ]: _create a DigitalObject_<x_0.3122><y_0.4518><table><tr><td>item_1 = DigitalObject('data/workshop-6/item_1/')</tz><tr><td>print(f'DigitalObject name: {item_1.name}')</tz><tr><td>print(f'number of directories: {len(item_1.dir_list)}')</tz><tr><td>print(f'number of files: {len(item_1.files_list)}')</tz><tr><td>In [ ]: _comparison_\n\ning item_1.extra_files_list_total > 0:<x_0.4067><y_0.4912>if item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > 0:\n\nif item_1.extra_files_list_total > extra files')<x_0.5067><y_0.5276>if item_1.extra_files_list_total === 1:\n\n>else:<x_0.1752><y_0.5336>mini('1 extra file')\n\n>print(f'All Extra Files: {[x.name for x in item_1.extra_files_list]}')\n\nrange: connector=double_arrow<y_0.7091>overlay_1.scatter><y_0.7334><class_Text>\n\n<x_0.0706><y_0.7482>infogramed by each<y_0.8334><y_0.8412>function, which is the smallest of the system\n\nfit to the right and bottom of the system\n\n<table><y_0.9035><table><y_1.0042><table><y_1.1061><table><y_1.2084><y_1.3092><y_1.40046><y_1.5017><class_Table>\n\n<x_0.0706><y_0.9496><table><y_1.5032><y_1.60346><y_1.6142><class_Text>\n\n<x_0.0706><y_0.9528><y_1.7043><class_Text>", "job_id": null}, "started_at": "2026-10-31T12:47:40.072562", "completed_at": "2026-10-31T12:47:40.072562", "latency_in_ms": 25781}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__lightblue.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__lightblue", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__lightblue.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |\n\n| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |\n\n| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |"}, "output": {"task_type": "parse", "example_id": "text/text_misc__lightblue", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |", "line_numbers": []}, {"page_index": 0, "markdown": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |\n\n| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:21.~~928752~~", "completed_at": "2026-10-~~31T03~~:45:21.~~928752~~", "latency_in_ms": 1451}

+ {"request": {"example_id": "text/text_misc__lightblue", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__lightblue.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |\n\n| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |\n\n| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |"}, "output": {"task_type": "parse", "example_id": "text/text_misc__lightblue", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |", "line_numbers": []}, {"page_index": 0, "markdown": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |\n\n| Category | Rate |\n|---|---|\n| We could design a pollution detector that would simulate what the beluga whales do to measure pollution in the oceans. | 25938 |\n| column_1 | 25865 |\n| column_2 | 24593 |\n| column_3 | 23816 |", "job_id": null}, "started_at": "2026-10-31T12:47:40.425948", "completed_at": "2026-10-31T12:47:40.425948", "latency_in_ms": 1451}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__mangle.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__mangle", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__mangle.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.\n\n# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.\n\n# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students."}, "output": {"task_type": "parse", "example_id": "text/text_misc__mangle", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.", "line_numbers": []}, {"page_index": 0, "markdown": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.\n\n# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:22.~~285634~~", "completed_at": "2026-10-~~31T03~~:45:22.~~285634~~", "latency_in_ms": 106399}

+ {"request": {"example_id": "text/text_misc__mangle", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__mangle.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.\n\n# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.\n\n# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students."}, "output": {"task_type": "parse", "example_id": "text/text_misc__mangle", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.", "line_numbers": []}, {"page_index": 0, "markdown": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.\n\n# Common Data Set 2024-2025<br>\nUpdated: 09.16.2025\n\nTalent/ability (as admission factor): Special consideration given to students with demonstrated talent/abilities in areas of interest to the institution (e.g., sports, the arts, languages, etc.).<x_0.8261><y_0.1061>\nTeacher certification program: Program designed to prepare students to meet the requirements for certification as teachers in elementary, middle/junior high, and secondary schools.\n\nTransfer student: A student entering the institution for the first time but known to have previously attended a postsecondary institution at the same level (e.g., undergraduate). The student may transfer with or without credit.<x_0.9276><y_0.1248>\nTransportation (costs): Assume two round trips to student\u2019s hometown per year for students in institutional housing or daily travel to and from your institution for commuter students.", "job_id": null}, "started_at": "2026-10-31T12:47:40.787167", "completed_at": "2026-10-31T12:47:40.787167", "latency_in_ms": 106399}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__mark2.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__mark2", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__mark2.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>\n\n# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>\n\n# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>"}, "output": {"task_type": "parse", "example_id": "text/text_misc__mark2", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>", "line_numbers": []}, {"page_index": 0, "markdown": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>\n\n# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>", "job_id": null}, "started_at": "2026-10-31T03:45:22.623175", "completed_at": "2026-10-31T03:45:22.623175", "latency_in_ms": 30657}

+ {"request": {"example_id": "text/text_misc__mark2", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__mark2.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>\n\n# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>\n\n# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>"}, "output": {"task_type": "parse", "example_id": "text/text_misc__mark2", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>", "line_numbers": []}, {"page_index": 0, "markdown": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>\n\n# ACUITY\n\n<table><tr><th rowspan=\"2\"></th><th colspan=\"3\">ACUITY</th><th colspan=\"2\">Road and Residence</th><th>Rate and Rule Filing Manual</th></tr><tr><td>02-2012</td><td>Edition</td></tr><tr><td>MEDICAL PAYMENTS</td></tr><tr><td rowspan=\"4\"><td>$ 1. Medical Payments (Private Passenger Car, Utility Car, Special Interest Car)</td><td>a. Base rate <sup>t territory factor</sup></td></td></tr><tr><td>b. a. <sup>medical payment symbol factor</sup></td></tr><tr><td>c. b. <sup>liability model year factor (not applicable if nonowned auto)</sup></td></tr><tr><td>d. c. <sup>increased limit factor</sup></td></tr><tr><td>e. d. <sup>level factor</sup></td></tr><tr><td>f. e. <sup>responsibility factor</sup></td></tr><tr><td>g. f. <sup>valued customer factor</sup></td></tr><tr><td>h. g. <sup>safe driver discount factor</sup></td></tr><tr><td>i. h. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>j. i. <sup>persistency discount factor</sup></td></tr><tr><td>k. j. accident prevention course discount</td></tr><tr><td>l. k. <sup>point surcharge factor</sup></td></tr><tr><td>m. l. class factor</td></tr><tr><td>n. m.<td>driver age factors</td></td></tr><tr><td>o. n. <sup>annual mileage factors</sup></td></tr><tr><td>p. o. <sup>permissive use factor</sup></td></tr><tr><td>q. p. <sup>nonowned discount factor (if applicable)</sup></td></tr><tr><td>r. q. <sup>term factor</sup></td></tr><tr><td>s. r. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>t. s. renewal premium cap factor (if applicable)</td></tr><tr><td rowspan=\"2\">Rounding: Round step t. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr><tr><td>2. Medical Payments (Antique Car)</td></tr><tr><td>a. Base rate <sup>t territory factor</sup></td></tr><tr><td>b. a. <sup>increased limit factor</sup></td></tr><tr><td>c. b. <sup>level factor</sup></td></tr><tr><td>d. c. <sup>responsibility factor</sup></td></tr><tr><td>e. d. <sup>valued customer factor</sup></td></tr><tr><td>f. e. <sup>safe driver discount factor</sup></td></tr><tr><td>g. f. <sup>reinstatement surcharge factor</sup></td></tr><tr><td>h. g. <sup>persistency discount factor</sup></td></tr><tr><td>i. h. <sup>class factor</sup></td></tr><tr><td>j. i. .25</td></tr><tr><td>k. j. term factor</td></tr><tr><td>l. k. <sup>profession discount factor (if applicable)</sup></td></tr><tr><td>m. l. <sup>renewal premium cap factor (if applicable)</sup></td></tr><tr><td rowspan=\"2\">Round: Round step m. to the nearest 1.00. Round all other steps to the nearest .10.</td></tr></table>\n\n<table><y_0.9679><class_Table>\n\n<class_Text>\n\n<x_0.1709><y_0.2420<class_Text>\n\n<x_0.1709><y_0.2530<class_Text>", "job_id": null}, "started_at": "2026-10-31T12:47:41.132905", "completed_at": "2026-10-31T12:47:41.132905", "latency_in_ms": 30657}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__marks.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__marks", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__marks.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |\n\n# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |\n\n# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |"}, "output": {"task_type": "parse", "example_id": "text/text_misc__marks", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |", "line_numbers": []}, {"page_index": 0, "markdown": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |\n\n# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |", "job_id": null}, "started_at": "2026-10-31T03:45:23.055670", "completed_at": "2026-10-31T03:45:23.055670", "latency_in_ms": 106272}

+ {"request": {"example_id": "text/text_misc__marks", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__marks.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |\n\n# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |\n\n# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |"}, "output": {"task_type": "parse", "example_id": "text/text_misc__marks", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |", "line_numbers": []}, {"page_index": 0, "markdown": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |\n\n# | category | percentage |\n|---|---|\n| series_15 | ASSOCIATED CONTENT |\n| segment_20 | Maroc 21 | Maroc 23 |\n| segment_25 | Maroc 24 | Maroc 26 |\n| segment_30 | Maroc 28 | Maroc 30 |\n| segment_35 | Maroc 32 | Maroc 34 |\n| segment_40 | Maroc 36 | Maroc 40 |\n| segment_45 | Maroc 48 | Maroc 49 |\n| segment_50 | Maroc 50 | Maroc 52 |\n| segment_55 | Maroc 56 | Maroc 58 |\n| segment_60 | Maroc 60 | Maroc 62 |\n| segment_65 | Maroc 64 | Maroc 65 |\n| segment_70 | Maroc 66 | Maroc 68 |\n| segment_75 | Maroc 70 | Maroc 72 |\n| segment_80 | Maroc 72 | Maroc 74 |\n| segment_85 | Maroc 80 | Maroc 76 |\n| segment_90 | Maroc 85 | Maroc 88 |\n| segment_95 | Maroc 90 | Maroc 92 |\n| segment_100 | Maroc 100 | Maroc 105 |\n| segment_105 | Maroc 110 | Maroc 120 |\n| segment_110 | Maroc 120 | Maroc 130 |\n| segment_120 | Maroc 130 | Maroc 140 |\n| segment_130 | Maroc 140 | Maroc 150 |\n| segment_140 | Maroc 150 | Maroc 160 |\n| segment_150 | Maroc 160 | Maroc 170 |\n| segment_160 | Maroc 170 | Maroc 180 |\n| segment_170 | Maroc 180 | Maroc 190 |\n| segment_180 | Maroc 190 | Maroc 200 |\n| segment_190 | Maroc 210 | Maroc 220 |\n| segment_200 | Maroc 220 | Maroc 230 |\n| segment_2100 | Maroc 230 | Maroc 240 |\n| segment_2200 | Maroc 240 | Maroc 250 |\n| segment_2300 | Maroc 250 | Maroc 260 |\n| segment_2400 | Maroc 260 | Maroc 270 |\n| segment_2500 | Maroc 270 | Maroc 280 |\n| segment_2600 | Maroc 280 | Maroc 290 |\n| segment_2700 | Maroc 290 | Maroc 300 |\n| segment_2800 | Maroc 310 | Maroc 320 |\n| segment_2900 | Maroc 320 | Maroc 330 |\n| segment_3000 | Maroc 340 | Maroc 350 |\n| segment_3100 | Maroc 350 | Maroc 360 |\n| segment_3200 | Maroc 360 | Maroc 370 |\n| segment_3300 | Maroc 380 | Maroc 390 |\n| segment_3400 | Maroc 390 | Maroc 400 |\n| segment_3500 | Maroc 4100 | Maroc 420 |\n| segment_3600 | Maroc 4200 | Maroc 430 |\n| segment_3700 | Maroc 4400 | Maroc 450 |\n| segment_3800 | Maroc 4500 | Maroc 460 |\n| segment_3900 | Maroc 4600 | Maroc 470 |\n| segment_4000 | Maroc 4800 | Maroc 490 |\n| segment_41000 | Maroc 5000 | Maroc 5100 |\n| segment_42000 | Maroc 5200 | Maroc 5300 |\n| segment_43000 | Maroc 5400 | Maroc 5500 |\n| segment_44000 | Maroc 5600 | Maroc 5600 |\n| segment_45000 | Maroc 5700 | Maroc 5800 |\n| segment_46000 | Maroc 5900 | Maroc 6000 |\n| segment_47000 | Maroc 6000 | Maroc 6100 |\n| segment_48000 | Maroc 6200 | Maroc 6300 |\n| segment_49000 | Maroc 64000 | Maroc 6500 |\n| segment_50000 | Maroc 65000 | Maroc 6600 |\n| segment_510000 | Maroc 66000 | Maroc 6700 |\n| segment_520000 | Maroc 68000 | Maroc 6900 |\n| segment_530000 | Maroc 69000 | Maroc 7000 |\n| segment_540000 | Maroc 70000 | Maroc 7100 |\n| segment_550000 | Maroc 72000 | Maroc 7200 |\n| segment_560000 | Maroc 73000 | Maroc 7400 |\n| segment_570000 | Maroc 750000 | Maroc 7600 |\n| segment_580000 | Maroc 760000 | Maroc 7700 |\n| segment_590000 | Maroc 780000 | Maroc 7900 |\n| segment_600000 | Maroc 790000 | Maroc 8000 |\n| segment_6100000 | Maroc 8100000 | Maroc 82000 |\n| segment_6200000 | Maroc 8300000 | Maroc 840000 |\n\n<table[7]:\n| Asset_10 | Asset_20 | Asset_30 | Asset_40 |\n| segment_20 | Asset_30 | Asset_45 |\n| segment_30 | Asset_45 | Asset_50 |\n| segment_40 | Asset_50 | Asset_60 |\n| segment_50 | Asset_60 | Asset_70 |\n| segment_60 | Asset_70 | Asset_80 |\n| segment_70 | Asset_80 | Asset_90 |\n| segment_80 | Asset_90 | Asset_950 |\n| segment_90 | Asset_100 | Asset_110 |\n| segment_100 | Asset_120 | Asset_13000 |", "job_id": null}, "started_at": "2026-10-31T12:47:41.539376", "completed_at": "2026-10-31T12:47:41.539376", "latency_in_ms": 106272}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__missing.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__notATable.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__overlapping.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__overlapping", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__overlapping.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "**LangCheck**\n\n**LangCheck**", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "**LangCheck**\n\n**LangCheck**"}, "output": {"task_type": "parse", "example_id": "text/text_misc__overlapping", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "**LangCheck**", "line_numbers": []}, {"page_index": 0, "markdown": "**LangCheck**", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "**LangCheck**\n\n**LangCheck**", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:24.~~270379~~", "completed_at": "2026-10-~~31T03~~:45:24.~~270379~~", "latency_in_ms": 50437}

+ {"request": {"example_id": "text/text_misc__overlapping", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__overlapping.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "**LangCheck**\n\n**LangCheck**", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "**LangCheck**\n\n**LangCheck**"}, "output": {"task_type": "parse", "example_id": "text/text_misc__overlapping", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "**LangCheck**", "line_numbers": []}, {"page_index": 0, "markdown": "**LangCheck**", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "**LangCheck**\n\n**LangCheck**", "job_id": null}, "started_at": "2026-10-31T12:47:42.758224", "completed_at": "2026-10-31T12:47:42.758224", "latency_in_ms": 50437}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__redacted.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__redacted", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__redacted.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM\n\n| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM\n\n| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM"}, "output": {"task_type": "parse", "example_id": "text/text_misc__redacted", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM", "line_numbers": []}, {"page_index": 0, "markdown": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM\n\n| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM", "job_id": null}, "started_at": "2026-10-31T03:45:24.677256", "completed_at": "2026-10-31T03:45:24.677256", "latency_in_ms": 15031}

+ {"request": {"example_id": "text/text_misc__redacted", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__redacted.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM\n\n| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM\n\n| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM"}, "output": {"task_type": "parse", "example_id": "text/text_misc__redacted", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM", "line_numbers": []}, {"page_index": 0, "markdown": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM\n\n| From: Bickerstaffe, Keir - SOL [/O=EXCHANGELABS/OU=EXCHANGE ADMINISTRATIVE GROUP (FYDIBOHF23SPDLT)/CN=RECIPIENTS/CN=76946342408840629D14D5B8C536A764-BICKERSTAFF]<x_0.8351><y_0.0661>\nSent: 10/8/2020 2:26:40 PM\n\n<class_Text>\n\nccccccc}\n**Subject:** & **Subject:** & RE: EO 13950 questions<x_0.3170><y_0.1067><y_0.3203><class_Text>\n\n<x_0.0586><sup>Subject</sup>: NILG EO 13950 Presentation.pdf\n\n# As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n## <class_Text>\n\n<x_0.0596><y_0.3836>All,<x_0.0706><y_0.1433>As I mentioned in our meeting this afternoon, attached is a copy of the slide deck used at NILG\u2019s webinar yesterday on EO 13950.\n\n<class_Text>\n\n<sup>(202)</sup><y_0.1832><class_Text>\n\n<sup>(b) 6</sup></sup><y_0.1855>From: Leen, Craig - OFCCP <Leen.Craig@DOL.gov>\n\nSent: Wednesday, October 7, 2020 4:14 PM\n\nTo: Dankowitz, Beverly - SOL (b) 6 @dol.gov>; Bickerstaffe, Keir - SOL (b) 6 @dol.gov>; Lupardo, Jeffrey - SOL (b) 6 @dol.gov>\n\nCc: Gaglione, Robert J - OFCCP <Gaglione.Robert.J@dol.gov>; Davidson, Patricia J - OFCCP (b) 6 @dol.gov><y_0.1542><y_0.6375><class_Text>\n\n<x_0.0596><y_0.6367>Williams, Tina T - OFCCP (b) 6 @dol.gov>; Gean, Lissette - OFCCP (b) 6 @dol.gov><y_0.2738><class_Text>\n\n<sup>Subject</sup>: RE: EO 13950 questions\n\n<class_Text>\n\n<sup>Craig E. Leen</x_0.1352><y_0.3520><class_Text>\n\n<sup>FOCCP Director</x_0.1678><y_0.3670><class_Text>\n\n<sup>100</sup>\n\n<class_Text>\n\n<sup>Sent:</sup> Wednesday, October 07, 2020 4:14 PM", "job_id": null}, "started_at": "2026-10-31T12:47:43.180336", "completed_at": "2026-10-31T12:47:43.180336", "latency_in_ms": 15031}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__redacted3.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__redacted3", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__redacted3.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |\n\n# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |\n\n# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |"}, "output": {"task_type": "parse", "example_id": "text/text_misc__redacted3", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |", "line_numbers": []}, {"page_index": 0, "markdown": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |\n\n# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |", "job_id": null}, "started_at": "2026-10-31T03:45:25.100065", "completed_at": "2026-10-31T03:45:25.100065", "latency_in_ms": 12286}

+ {"request": {"example_id": "text/text_misc__redacted3", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__redacted3.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |\n\n# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |\n\n# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |"}, "output": {"task_type": "parse", "example_id": "text/text_misc__redacted3", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |", "line_numbers": []}, {"page_index": 0, "markdown": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |\n\n# FW: Final Version of Spring 2018 Agenda for Submission into\nROCIS and Signature\n\nFrom: \"Neumayr, Mary B. EOP/CEQ\" <b) (6)<y_0.1061><class_Text>\n\n<x_0.0706><y_0.1176>\"To: \"Szabo, Aaron L. EOP/CEQ\" <b) (6)<y_0.1184>| \"Smith, Katherine R.\n\nCc: \"Seale, Viktoria Z. EOP/CEQ\" <b) (6)<y_0.1396><class_Text>\n\n<x_0.0706><y_0.1448>Date: Mon, 05 Mar 2018 16:51:38 -0500\n\n## Attachments\n\nUPDATED CEQ Unified Agenda Entries--Spring 2018 Final.docx (26.91 kB)\n\nI have attached an updated draft with minor additional revisions. Please use this version. Thanks,\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality\n\n(b) (6)<y_0.1928>(office), (6)<y_0.2125>(cell)<y_0.2276><y_0.5047><class_Text>\n\n<x_0.0706><y_0.5266>From: Neumayr, Mary B. EOP/CEQ\n\nSent: Monday, March 5, 2018 4:35 PM\n\nTo: Szabo, Aaron L. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.3552>Seale, Viktoria Z. EOP/CEQ\n\nSubject: RE: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\nSuggested edits attached.\n\n## Mary B. Neumayr\n\nChief of Staff, Council on Environmental Quality<x_0.4078><y_0.3938><y_0.7094><class_Text>\n\n<x_0.0706><y_0.7102>(b) (6)<y_0.4152><y_0.7227><class_Text>\n\n<x_0.0706><y_0.7461>From: Szabo, Aaron L. EOP/CEQ\n\nSent: Monday, March 5, 2018 3:28 PM\n\nTo: Neumayr, Mary B. EOP/CEQ\n\nCc: Smith, Katherine R. EOP/CEQ\n\n<y_0.4332>\u2018Szabo, Aaron L. EOP/CEQ\n\nSubject: Final Version of Spring 2018 Agenda for Submission into ROCIS and Signature\n\n<y_0.4738><table>[7]><y_0.4912><table><y_0.5033><table><table>table[3]:\n| category_1 | column_2 |\n| Category | column_3 |\n| Category | column_3 |\n| Category | column_4 |\n| Source | column_4 |\n| Source | column_4 |", "job_id": null}, "started_at": "2026-10-31T12:47:43.597291", "completed_at": "2026-10-31T12:47:43.597291", "latency_in_ms": 12286}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__reverRo.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__skew3.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__skew3", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__skew3.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# -2- September 29, 2005\n\nWe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nwe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nOn February 27, 2002, the Court ruled that the provincial government and the developer did not properly consult the Council of the Haida Nation when renewing a tree farm licence. The Court further held that a legally enforceable duty to consult will continue until Haida title and rights are determined by a treaty or by a Court.<x_0.8379><y_0.1\nThe Haida decision followed a January 31, 2002 decision by the Court of Appeal in the Taku River Tlingit case. In that case, the Court ruled that the province must consult Taku River Tlingit First Nation before they re-issue a project permit to a developer to reopen the Tulsequah Chief Mine in northwestern B.C.<x_0.8379><y_0.1458><table><tr><th>Haida, the Court referred to the compelling case the Haida had to aboriginal title in the Queen Charlotte Islands as requiring the duty to consult. In Taku, the Court referred to the aboriginal interests expressed at the treaty table as evidence of Taku's interest.<y_0.1636><class_text>\n<x_0.0706><y_0.1734>We believe that these examples of recent Case Law are pertinent to our current situation. A development is being proposed in our traditional territory and we have not been consulted or included in any initial discussions regarding the potential impact said development may have on our traditional lands.\n\nWe note that during the period of preliminary application, Doublestar Resources replied to your office (April 6, 2005) referring to groups and individuals they had contacted following conversations with your office. Strangely, the North Slave M\u00e9tis Alliance was included on the list, while the NWTMN was not. It would appear that the proponent was not advised of our organization's existence, thereby rendering it impossible for that proponent to initiate discussions with us.\n\nIn light of the above, we have no option to but to request more time to consider the proposal and enter into discussions with the proponent regarding the proposed activities and what impacts such activities might have on our traditional territory.\n\n## The NWTMN bases its position on the grounds that:\n\nThe application indicates there is potential for environmental impacts. The NWTMN has not been informed of the nature and extent of environmental impacts as a result of the proponent\u2019s proposed activities in our traditional territory.\n\nThe NWTMN has not been informed of the cumulative loss to M\u00e9tis resources should this project go ahead.<x_0.8176><y_0.4676><y_0.8195><class_text>\n<x_0.0706><y_0.5048><table><y_0.5152><table><table>series</table></table></y_0.5152></table></y_0.5048></x_0.0706></class_text></y_0.8195></y_0.4676></x_0.8176></y_0.1734></x_0.0706></class_text></y_0.1636></th></tr></table><table>percentage=7990><table>series=1000000>| category | value |\n| --- | --- |\n| 29 2005 | 3.35 |\n| 86 78722772 | 8.75 |\n\nscan005.TIF scan000.TIF", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# -2- September 29, 2005\n\nWe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nwe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nOn February 27, 2002, the Court ruled that the provincial government and the developer did not properly consult the Council of the Haida Nation when renewing a tree farm licence. The Court further held that a legally enforceable duty to consult will continue until Haida title and rights are determined by a treaty or by a Court.<x_0.8379><y_0.1\nThe Haida decision followed a January 31, 2002 decision by the Court of Appeal in the Taku River Tlingit case. In that case, the Court ruled that the province must consult Taku River Tlingit First Nation before they re-issue a project permit to a developer to reopen the Tulsequah Chief Mine in northwestern B.C.<x_0.8379><y_0.1458><table><tr><th>Haida, the Court referred to the compelling case the Haida had to aboriginal title in the Queen Charlotte Islands as requiring the duty to consult. In Taku, the Court referred to the aboriginal interests expressed at the treaty table as evidence of Taku's interest.<y_0.1636><class_text>\n<x_0.0706><y_0.1734>We believe that these examples of recent Case Law are pertinent to our current situation. A development is being proposed in our traditional territory and we have not been consulted or included in any initial discussions regarding the potential impact said development may have on our traditional lands.\n\nWe note that during the period of preliminary application, Doublestar Resources replied to your office (April 6, 2005) referring to groups and individuals they had contacted following conversations with your office. Strangely, the North Slave M\u00e9tis Alliance was included on the list, while the NWTMN was not. It would appear that the proponent was not advised of our organization's existence, thereby rendering it impossible for that proponent to initiate discussions with us.\n\nIn light of the above, we have no option to but to request more time to consider the proposal and enter into discussions with the proponent regarding the proposed activities and what impacts such activities might have on our traditional territory.\n\n## The NWTMN bases its position on the grounds that:\n\nThe application indicates there is potential for environmental impacts. The NWTMN has not been informed of the nature and extent of environmental impacts as a result of the proponent\u2019s proposed activities in our traditional territory.\n\nThe NWTMN has not been informed of the cumulative loss to M\u00e9tis resources should this project go ahead.<x_0.8176><y_0.4676><y_0.8195><class_text>\n<x_0.0706><y_0.5048><table><y_0.5152><table><table>series</table></table></y_0.5152></table></y_0.5048></x_0.0706></class_text></y_0.8195></y_0.4676></x_0.8176></y_0.1734></x_0.0706></class_text></y_0.1636></th></tr></table><table>percentage=7990><table>series=1000000>| category | value |\n| --- | --- |\n| 29 2005 | 3.35 |\n| 86 78722772 | 8.75 |\n\nscan005.TIF scan000.TIF"}, "output": {"task_type": "parse", "example_id": "text/text_misc__skew3", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# -2- September 29, 2005\n\nWe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nwe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nOn February 27, 2002, the Court ruled that the provincial government and the developer did not properly consult the Council of the Haida Nation when renewing a tree farm licence. The Court further held that a legally enforceable duty to consult will continue until Haida title and rights are determined by a treaty or by a Court.<x_0.8379><y_0.1\nThe Haida decision followed a January 31, 2002 decision by the Court of Appeal in the Taku River Tlingit case. In that case, the Court ruled that the province must consult Taku River Tlingit First Nation before they re-issue a project permit to a developer to reopen the Tulsequah Chief Mine in northwestern B.C.<x_0.8379><y_0.1458><table><tr><th>Haida, the Court referred to the compelling case the Haida had to aboriginal title in the Queen Charlotte Islands as requiring the duty to consult. In Taku, the Court referred to the aboriginal interests expressed at the treaty table as evidence of Taku's interest.<y_0.1636><class_text>\n<x_0.0706><y_0.1734>We believe that these examples of recent Case Law are pertinent to our current situation. A development is being proposed in our traditional territory and we have not been consulted or included in any initial discussions regarding the potential impact said development may have on our traditional lands.\n\nWe note that during the period of preliminary application, Doublestar Resources replied to your office (April 6, 2005) referring to groups and individuals they had contacted following conversations with your office. Strangely, the North Slave M\u00e9tis Alliance was included on the list, while the NWTMN was not. It would appear that the proponent was not advised of our organization's existence, thereby rendering it impossible for that proponent to initiate discussions with us.\n\nIn light of the above, we have no option to but to request more time to consider the proposal and enter into discussions with the proponent regarding the proposed activities and what impacts such activities might have on our traditional territory.\n\n## The NWTMN bases its position on the grounds that:\n\nThe application indicates there is potential for environmental impacts. The NWTMN has not been informed of the nature and extent of environmental impacts as a result of the proponent\u2019s proposed activities in our traditional territory.\n\nThe NWTMN has not been informed of the cumulative loss to M\u00e9tis resources should this project go ahead.<x_0.8176><y_0.4676><y_0.8195><class_text>\n<x_0.0706><y_0.5048><table><y_0.5152><table><table>series</table></table></y_0.5152></table></y_0.5048></x_0.0706></class_text></y_0.8195></y_0.4676></x_0.8176></y_0.1734></x_0.0706></class_text></y_0.1636></th></tr></table><table>percentage=7990><table>series=1000000>| category | value |\n| --- | --- |\n| 29 2005 | 3.35 |\n| 86 78722772 | 8.75 |\n\nscan005.TIF scan000.TIF", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# -2- September 29, 2005\n\nWe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nwe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nOn February 27, 2002, the Court ruled that the provincial government and the developer did not properly consult the Council of the Haida Nation when renewing a tree farm licence. The Court further held that a legally enforceable duty to consult will continue until Haida title and rights are determined by a treaty or by a Court.<x_0.8379><y_0.1\nThe Haida decision followed a January 31, 2002 decision by the Court of Appeal in the Taku River Tlingit case. In that case, the Court ruled that the province must consult Taku River Tlingit First Nation before they re-issue a project permit to a developer to reopen the Tulsequah Chief Mine in northwestern B.C.<x_0.8379><y_0.1458><table><tr><th>Haida, the Court referred to the compelling case the Haida had to aboriginal title in the Queen Charlotte Islands as requiring the duty to consult. In Taku, the Court referred to the aboriginal interests expressed at the treaty table as evidence of Taku's interest.<y_0.1636><class_text>\n<x_0.0706><y_0.1734>We believe that these examples of recent Case Law are pertinent to our current situation. A development is being proposed in our traditional territory and we have not been consulted or included in any initial discussions regarding the potential impact said development may have on our traditional lands.\n\nWe note that during the period of preliminary application, Doublestar Resources replied to your office (April 6, 2005) referring to groups and individuals they had contacted following conversations with your office. Strangely, the North Slave M\u00e9tis Alliance was included on the list, while the NWTMN was not. It would appear that the proponent was not advised of our organization's existence, thereby rendering it impossible for that proponent to initiate discussions with us.\n\nIn light of the above, we have no option to but to request more time to consider the proposal and enter into discussions with the proponent regarding the proposed activities and what impacts such activities might have on our traditional territory.\n\n## The NWTMN bases its position on the grounds that:\n\nThe application indicates there is potential for environmental impacts. The NWTMN has not been informed of the nature and extent of environmental impacts as a result of the proponent\u2019s proposed activities in our traditional territory.\n\nThe NWTMN has not been informed of the cumulative loss to M\u00e9tis resources should this project go ahead.<x_0.8176><y_0.4676><y_0.8195><class_text>\n<x_0.0706><y_0.5048><table><y_0.5152><table><table>series</table></table></y_0.5152></table></y_0.5048></x_0.0706></class_text></y_0.8195></y_0.4676></x_0.8176></y_0.1734></x_0.0706></class_text></y_0.1636></th></tr></table><table>percentage=7990><table>series=1000000>| category | value |\n| --- | --- |\n| 29 2005 | 3.35 |\n| 86 78722772 | 8.75 |\n\nscan005.TIF scan000.TIF", "job_id": null}, "started_at": "2026-10-31T03:45:27.140885", "completed_at": "2026-10-31T03:45:27.140885", "latency_in_ms": 10749}

+ {"request": {"example_id": "text/text_misc__skew3", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__skew3.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# -2- September 29, 2005\n\nWe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nwe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nOn February 27, 2002, the Court ruled that the provincial government and the developer did not properly consult the Council of the Haida Nation when renewing a tree farm licence. The Court further held that a legally enforceable duty to consult will continue until Haida title and rights are determined by a treaty or by a Court.<x_0.8379><y_0.1\nThe Haida decision followed a January 31, 2002 decision by the Court of Appeal in the Taku River Tlingit case. In that case, the Court ruled that the province must consult Taku River Tlingit First Nation before they re-issue a project permit to a developer to reopen the Tulsequah Chief Mine in northwestern B.C.<x_0.8379><y_0.1458><table><tr><th>Haida, the Court referred to the compelling case the Haida had to aboriginal title in the Queen Charlotte Islands as requiring the duty to consult. In Taku, the Court referred to the aboriginal interests expressed at the treaty table as evidence of Taku's interest.<y_0.1636><class_text>\n<x_0.0706><y_0.1734>We believe that these examples of recent Case Law are pertinent to our current situation. A development is being proposed in our traditional territory and we have not been consulted or included in any initial discussions regarding the potential impact said development may have on our traditional lands.\n\nWe note that during the period of preliminary application, Doublestar Resources replied to your office (April 6, 2005) referring to groups and individuals they had contacted following conversations with your office. Strangely, the North Slave M\u00e9tis Alliance was included on the list, while the NWTMN was not. It would appear that the proponent was not advised of our organization's existence, thereby rendering it impossible for that proponent to initiate discussions with us.\n\nIn light of the above, we have no option to but to request more time to consider the proposal and enter into discussions with the proponent regarding the proposed activities and what impacts such activities might have on our traditional territory.\n\n## The NWTMN bases its position on the grounds that:\n\nThe application indicates there is potential for environmental impacts. The NWTMN has not been informed of the nature and extent of environmental impacts as a result of the proponent\u2019s proposed activities in our traditional territory.\n\nThe NWTMN has not been informed of the cumulative loss to M\u00e9tis resources should this project go ahead.<x_0.8176><y_0.4676><y_0.8195><class_text>\n<x_0.0706><y_0.5048><table><y_0.5152><table><table>series</table></table></y_0.5152></table></y_0.5048></x_0.0706></class_text></y_0.8195></y_0.4676></x_0.8176></y_0.1734></x_0.0706></class_text></y_0.1636></th></tr></table><table>percentage=7990><table>series=1000000>| category | value |\n| --- | --- |\n| 29 2005 | 3.35 |\n| 86 78722772 | 8.75 |\n\nscan005.TIF scan000.TIF", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# -2- September 29, 2005\n\nWe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nwe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nOn February 27, 2002, the Court ruled that the provincial government and the developer did not properly consult the Council of the Haida Nation when renewing a tree farm licence. The Court further held that a legally enforceable duty to consult will continue until Haida title and rights are determined by a treaty or by a Court.<x_0.8379><y_0.1\nThe Haida decision followed a January 31, 2002 decision by the Court of Appeal in the Taku River Tlingit case. In that case, the Court ruled that the province must consult Taku River Tlingit First Nation before they re-issue a project permit to a developer to reopen the Tulsequah Chief Mine in northwestern B.C.<x_0.8379><y_0.1458><table><tr><th>Haida, the Court referred to the compelling case the Haida had to aboriginal title in the Queen Charlotte Islands as requiring the duty to consult. In Taku, the Court referred to the aboriginal interests expressed at the treaty table as evidence of Taku's interest.<y_0.1636><class_text>\n<x_0.0706><y_0.1734>We believe that these examples of recent Case Law are pertinent to our current situation. A development is being proposed in our traditional territory and we have not been consulted or included in any initial discussions regarding the potential impact said development may have on our traditional lands.\n\nWe note that during the period of preliminary application, Doublestar Resources replied to your office (April 6, 2005) referring to groups and individuals they had contacted following conversations with your office. Strangely, the North Slave M\u00e9tis Alliance was included on the list, while the NWTMN was not. It would appear that the proponent was not advised of our organization's existence, thereby rendering it impossible for that proponent to initiate discussions with us.\n\nIn light of the above, we have no option to but to request more time to consider the proposal and enter into discussions with the proponent regarding the proposed activities and what impacts such activities might have on our traditional territory.\n\n## The NWTMN bases its position on the grounds that:\n\nThe application indicates there is potential for environmental impacts. The NWTMN has not been informed of the nature and extent of environmental impacts as a result of the proponent\u2019s proposed activities in our traditional territory.\n\nThe NWTMN has not been informed of the cumulative loss to M\u00e9tis resources should this project go ahead.<x_0.8176><y_0.4676><y_0.8195><class_text>\n<x_0.0706><y_0.5048><table><y_0.5152><table><table>series</table></table></y_0.5152></table></y_0.5048></x_0.0706></class_text></y_0.8195></y_0.4676></x_0.8176></y_0.1734></x_0.0706></class_text></y_0.1636></th></tr></table><table>percentage=7990><table>series=1000000>| category | value |\n| --- | --- |\n| 29 2005 | 3.35 |\n| 86 78722772 | 8.75 |\n\nscan005.TIF scan000.TIF"}, "output": {"task_type": "parse", "example_id": "text/text_misc__skew3", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# -2- September 29, 2005\n\nWe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nwe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nOn February 27, 2002, the Court ruled that the provincial government and the developer did not properly consult the Council of the Haida Nation when renewing a tree farm licence. The Court further held that a legally enforceable duty to consult will continue until Haida title and rights are determined by a treaty or by a Court.<x_0.8379><y_0.1\nThe Haida decision followed a January 31, 2002 decision by the Court of Appeal in the Taku River Tlingit case. In that case, the Court ruled that the province must consult Taku River Tlingit First Nation before they re-issue a project permit to a developer to reopen the Tulsequah Chief Mine in northwestern B.C.<x_0.8379><y_0.1458><table><tr><th>Haida, the Court referred to the compelling case the Haida had to aboriginal title in the Queen Charlotte Islands as requiring the duty to consult. In Taku, the Court referred to the aboriginal interests expressed at the treaty table as evidence of Taku's interest.<y_0.1636><class_text>\n<x_0.0706><y_0.1734>We believe that these examples of recent Case Law are pertinent to our current situation. A development is being proposed in our traditional territory and we have not been consulted or included in any initial discussions regarding the potential impact said development may have on our traditional lands.\n\nWe note that during the period of preliminary application, Doublestar Resources replied to your office (April 6, 2005) referring to groups and individuals they had contacted following conversations with your office. Strangely, the North Slave M\u00e9tis Alliance was included on the list, while the NWTMN was not. It would appear that the proponent was not advised of our organization's existence, thereby rendering it impossible for that proponent to initiate discussions with us.\n\nIn light of the above, we have no option to but to request more time to consider the proposal and enter into discussions with the proponent regarding the proposed activities and what impacts such activities might have on our traditional territory.\n\n## The NWTMN bases its position on the grounds that:\n\nThe application indicates there is potential for environmental impacts. The NWTMN has not been informed of the nature and extent of environmental impacts as a result of the proponent\u2019s proposed activities in our traditional territory.\n\nThe NWTMN has not been informed of the cumulative loss to M\u00e9tis resources should this project go ahead.<x_0.8176><y_0.4676><y_0.8195><class_text>\n<x_0.0706><y_0.5048><table><y_0.5152><table><table>series</table></table></y_0.5152></table></y_0.5048></x_0.0706></class_text></y_0.8195></y_0.4676></x_0.8176></y_0.1734></x_0.0706></class_text></y_0.1636></th></tr></table><table>percentage=7990><table>series=1000000>| category | value |\n| --- | --- |\n| 29 2005 | 3.35 |\n| 86 78722772 | 8.75 |\n\nscan005.TIF scan000.TIF", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# -2- September 29, 2005\n\nWe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nwe take this opportunity to reiterate your definition of consultation in light of two recent landmark rulings in the BC Court of Appeal that held that the provincial government (Crown) has a duty to properly consult First Nations before proceeding with development on potential treaty settlement land. (It should be noted here that the NWT M\u00e9tis Nation falls under the definition of \"first nation\" in the MVRMA under the heading of \"interpretation.\")\n\nOn February 27, 2002, the Court ruled that the provincial government and the developer did not properly consult the Council of the Haida Nation when renewing a tree farm licence. The Court further held that a legally enforceable duty to consult will continue until Haida title and rights are determined by a treaty or by a Court.<x_0.8379><y_0.1\nThe Haida decision followed a January 31, 2002 decision by the Court of Appeal in the Taku River Tlingit case. In that case, the Court ruled that the province must consult Taku River Tlingit First Nation before they re-issue a project permit to a developer to reopen the Tulsequah Chief Mine in northwestern B.C.<x_0.8379><y_0.1458><table><tr><th>Haida, the Court referred to the compelling case the Haida had to aboriginal title in the Queen Charlotte Islands as requiring the duty to consult. In Taku, the Court referred to the aboriginal interests expressed at the treaty table as evidence of Taku's interest.<y_0.1636><class_text>\n<x_0.0706><y_0.1734>We believe that these examples of recent Case Law are pertinent to our current situation. A development is being proposed in our traditional territory and we have not been consulted or included in any initial discussions regarding the potential impact said development may have on our traditional lands.\n\nWe note that during the period of preliminary application, Doublestar Resources replied to your office (April 6, 2005) referring to groups and individuals they had contacted following conversations with your office. Strangely, the North Slave M\u00e9tis Alliance was included on the list, while the NWTMN was not. It would appear that the proponent was not advised of our organization's existence, thereby rendering it impossible for that proponent to initiate discussions with us.\n\nIn light of the above, we have no option to but to request more time to consider the proposal and enter into discussions with the proponent regarding the proposed activities and what impacts such activities might have on our traditional territory.\n\n## The NWTMN bases its position on the grounds that:\n\nThe application indicates there is potential for environmental impacts. The NWTMN has not been informed of the nature and extent of environmental impacts as a result of the proponent\u2019s proposed activities in our traditional territory.\n\nThe NWTMN has not been informed of the cumulative loss to M\u00e9tis resources should this project go ahead.<x_0.8176><y_0.4676><y_0.8195><class_text>\n<x_0.0706><y_0.5048><table><y_0.5152><table><table>series</table></table></y_0.5152></table></y_0.5048></x_0.0706></class_text></y_0.8195></y_0.4676></x_0.8176></y_0.1734></x_0.0706></class_text></y_0.1636></th></tr></table><table>percentage=7990><table>series=1000000>| category | value |\n| --- | --- |\n| 29 2005 | 3.35 |\n| 86 78722772 | 8.75 |\n\nscan005.TIF scan000.TIF", "job_id": null}, "started_at": "2026-10-31T12:47:45.678798", "completed_at": "2026-10-31T12:47:45.678798", "latency_in_ms": 10749}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__stepbystep.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__stepbystep", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__stepbystep.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**\n\n# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**\n\n# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**"}, "output": {"task_type": "parse", "example_id": "text/text_misc__stepbystep", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**", "line_numbers": []}, {"page_index": 0, "markdown": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**\n\n# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:27.~~551074~~", "completed_at": "2026-10-~~31T03~~:45:27.~~551074~~", "latency_in_ms": 106632}

+ {"request": {"example_id": "text/text_misc__stepbystep", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__stepbystep.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**\n\n# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**\n\n# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**"}, "output": {"task_type": "parse", "example_id": "text/text_misc__stepbystep", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**", "line_numbers": []}, {"page_index": 0, "markdown": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**\n\n# **The Infant-Toddler Program Step-by-Step**\n\n**An overview of the steps your family may take while in the Infant-Toddler Program.**\n\n## **FIRST STEPS:**\n**GETTING STARTED**", "job_id": null}, "started_at": "2026-10-31T12:47:46.106855", "completed_at": "2026-10-31T12:47:46.106855", "latency_in_ms": 106632}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__templated.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__templated", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__templated.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]\n\n# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]\n\n# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]"}, "output": {"task_type": "parse", "example_id": "text/text_misc__templated", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]", "line_numbers": []}, {"page_index": 0, "markdown": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]\n\n# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]", "job_id": null}, "started_at": "2026-10-31T03:45:27.900047", "completed_at": "2026-10-31T03:45:27.900047", "latency_in_ms": 7942}

+ {"request": {"example_id": "text/text_misc__templated", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__templated.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]\n\n# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]\n\n# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]"}, "output": {"task_type": "parse", "example_id": "text/text_misc__templated", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]", "line_numbers": []}, {"page_index": 0, "markdown": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]\n\n# Illinois Department of Insurance\n\n## Letter of Submission Template\n\n[Issuer shall file a letter of submission, or provide information in the Filing Description field under the General Information tab in SERFF. (Reference 50 IAC 916.40(b) and 50 IAC 2001.130(a)(3))]<y_0.1709><class_Text>\n\n<x_0.0706><y_0.1857>1. [_Salutation_]\n\nRe: [_Issuer preference_]\n\n_Statement to indicate: (a) purpose and content of filing; (b) whether or not the submission is a new form_]\n\n4. [_List name of the form(s) and identifying form numbers. If the form is intended to supersede another, the form number of the form replaced and the date it was approved by the Department, with a redline version from the previously approved form. Identify SERFF tracking number(s) of the prior approved forms._]\n\n5. This is an [_insert one_ statement as shown below. Use all caps as shown._]\n\nON EXCHANGE filing. [_QHPs and SADPs. For a QHP form filing, indicate the SERFF tracking number of the associated Off Exchange mirrored form filing (a filing to represent an off exchange QHP plan that is also sold on the marketplace exchange)_]\n\n<y_0.3452>OFF EXCHANGE MIRRORED filing. [_QHPs only. Indicate the SERFF tracking number of the associated On Exchange form filing_]\n\nOFF EXCHANGE MARKET filing. [QHPs only. For forms/plans not offered on the marketplace exchange]\n\nOFF EXCHANGE CERTIFIED filing. [_SADPs only. For certified SADPs sold off exchange only_]\n\n6. [_Indicate the SERFF tracking number(s) of the associated rate, network, and form filings._]\n\n7. [_If applicable, statement to indicate whether or not the filing represents a Multi-State plan._]\n\n8. [_Other additional information, as appropriate for the filing._]\n\n9. [_Complimentary close/Name of Sender/Contact information for Sender_]", "job_id": null}, "started_at": "2026-10-31T12:47:46.455890", "completed_at": "2026-10-31T12:47:46.455890", "latency_in_ms": 7942}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_misc__yellow.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_misc__yellow", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__yellow.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33\n\n# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33\n\n# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33"}, "output": {"task_type": "parse", "example_id": "text/text_misc__yellow", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33", "line_numbers": []}, {"page_index": 0, "markdown": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33\n\n# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33", "job_id": null}, "started_at": "2026-10-31T03:45:28.429432", "completed_at": "2026-10-31T03:45:28.429432", "latency_in_ms": 6271}

+ {"request": {"example_id": "text/text_misc__yellow", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_misc__yellow.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33\n\n# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33\n\n# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33"}, "output": {"task_type": "parse", "example_id": "text/text_misc__yellow", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33", "line_numbers": []}, {"page_index": 0, "markdown": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33\n\n# Basic information about the observations\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\nFilament in the NOAA 12159 active region was observed on\n11 Sept 2014 9:28 \u2013 10:04 UT at VTT at Tenerife with Eche-\nlle spectrograph together with TIP1 spectropolarimeter in\nIR and simultaneously in H\u03b1 by the TESOS Farbry-P\u00e9rot\ninterferometer:\n\n\u2013 position of the filament: 215 arcsec, -420 arcsec; 3 scans\nmade with spectrograph between 9:28 \u2013 10:04 UT\n\nspectropolarimetric observations in chromospheric HeI\nIR triplet with wavelength around 10830 \u00c5, photospheric\nline SiI 10827 \u00c5; height of the slit approx. 30\", 1 pxl along\nthe slit 0.35\", scan step of 0.45\", scanned approx. 10\"\n\nspectroscopic observations in CaII 8542 \u00c5\n\n2D observations with the TESOS Fabry-P\u00e9rot interfero-\nmeter, FOV of 25''\u00d725\", step 0.025 \u00c5 of scanning in \u03bb\n\nwithin interval 6561.2 \u2013 6564.5 \u00c5\n\nthe 3rd scan was used, and TESOS observation No.33", "job_id": null}, "started_at": "2026-10-31T12:47:47.011243", "completed_at": "2026-10-31T12:47:47.011243", "latency_in_ms": 6271}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__10k2col.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_multicolumns__10k2col", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__10k2col.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "significantly change the competitive environment in which we conduct business. The financial services industry is also becoming more competitive as further technological advances enable more companies to provide financial products and services, including electronic and internet-based financial solutions such as electronic securities trading, lending and payment solutions, as well as digital currencies and alternative payment methods. These technological advances may diminish the importance of depository institutions and other financial intermediaries in the transfer of funds between parties.", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "significantly change the competitive environment in which we conduct business. The financial services industry is also becoming more competitive as further technological advances enable more companies to provide financial products and services, including electronic and internet-based financial solutions such as electronic securities trading, lending and payment solutions, as well as digital currencies and alternative payment methods. These technological advances may diminish the importance of depository institutions and other financial intermediaries in the transfer of funds between parties."}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__10k2col", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "significantly change the competitive environment in which we conduct business. The financial services industry is also becoming more competitive as further technological advances enable more companies to provide financial products and services, including electronic and internet-based financial solutions such as electronic securities trading, lending and payment solutions, as well as digital currencies and alternative payment methods. These technological advances may diminish the importance of depository institutions and other financial intermediaries in the transfer of funds between parties.", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "significantly change the competitive environment in which we conduct business. The financial services industry is also becoming more competitive as further technological advances enable more companies to provide financial products and services, including electronic and internet-based financial solutions such as electronic securities trading, lending and payment solutions, as well as digital currencies and alternative payment methods. These technological advances may diminish the importance of depository institutions and other financial intermediaries in the transfer of funds between parties.", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:28.~~776517~~", "completed_at": "2026-10-~~31T03~~:45:28.~~776517~~", "latency_in_ms": 106200}

+ {"request": {"example_id": "text/text_multicolumns__10k2col", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__10k2col.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "significantly change the competitive environment in which we conduct business. The financial services industry is also becoming more competitive as further technological advances enable more companies to provide financial products and services, including electronic and internet-based financial solutions such as electronic securities trading, lending and payment solutions, as well as digital currencies and alternative payment methods. These technological advances may diminish the importance of depository institutions and other financial intermediaries in the transfer of funds between parties.", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "significantly change the competitive environment in which we conduct business. The financial services industry is also becoming more competitive as further technological advances enable more companies to provide financial products and services, including electronic and internet-based financial solutions such as electronic securities trading, lending and payment solutions, as well as digital currencies and alternative payment methods. These technological advances may diminish the importance of depository institutions and other financial intermediaries in the transfer of funds between parties."}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__10k2col", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "significantly change the competitive environment in which we conduct business. The financial services industry is also becoming more competitive as further technological advances enable more companies to provide financial products and services, including electronic and internet-based financial solutions such as electronic securities trading, lending and payment solutions, as well as digital currencies and alternative payment methods. These technological advances may diminish the importance of depository institutions and other financial intermediaries in the transfer of funds between parties.", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "significantly change the competitive environment in which we conduct business. The financial services industry is also becoming more competitive as further technological advances enable more companies to provide financial products and services, including electronic and internet-based financial solutions such as electronic securities trading, lending and payment solutions, as well as digital currencies and alternative payment methods. These technological advances may diminish the importance of depository institutions and other financial intermediaries in the transfer of funds between parties.", "job_id": null}, "started_at": "2026-10-31T12:47:47.355605", "completed_at": "2026-10-31T12:47:47.355605", "latency_in_ms": 106200}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__14999.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_multicolumns__14999", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__14999.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Federal Register/Vol. 90, No. 65/Monday, April 7, 2025/Notices\n\nThe EA will be issued and circulated\nfor review by all interested parties. All\ncomments filed on the EA will be\nanalyzed by staff and considered in the\nCommission\u2019s final licensing decision.", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Federal Register/Vol. 90, No. 65/Monday, April 7, 2025/Notices\n\nThe EA will be issued and circulated\nfor review by all interested parties. All\ncomments filed on the EA will be\nanalyzed by staff and considered in the\nCommission\u2019s final licensing decision."}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__14999", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Federal Register/Vol. 90, No. 65/Monday, April 7, 2025/Notices\n\nThe EA will be issued and circulated\nfor review by all interested parties. All\ncomments filed on the EA will be\nanalyzed by staff and considered in the\nCommission\u2019s final licensing decision.", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Federal Register/Vol. 90, No. 65/Monday, April 7, 2025/Notices\n\nThe EA will be issued and circulated\nfor review by all interested parties. All\ncomments filed on the EA will be\nanalyzed by staff and considered in the\nCommission\u2019s final licensing decision.", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:29.~~122513~~", "completed_at": "2026-10-~~31T03~~:45:29.~~122513~~", "latency_in_ms": 106048}

+ {"request": {"example_id": "text/text_multicolumns__14999", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__14999.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Federal Register/Vol. 90, No. 65/Monday, April 7, 2025/Notices\n\nThe EA will be issued and circulated\nfor review by all interested parties. All\ncomments filed on the EA will be\nanalyzed by staff and considered in the\nCommission\u2019s final licensing decision.", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Federal Register/Vol. 90, No. 65/Monday, April 7, 2025/Notices\n\nThe EA will be issued and circulated\nfor review by all interested parties. All\ncomments filed on the EA will be\nanalyzed by staff and considered in the\nCommission\u2019s final licensing decision."}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__14999", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Federal Register/Vol. 90, No. 65/Monday, April 7, 2025/Notices\n\nThe EA will be issued and circulated\nfor review by all interested parties. All\ncomments filed on the EA will be\nanalyzed by staff and considered in the\nCommission\u2019s final licensing decision.", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Federal Register/Vol. 90, No. 65/Monday, April 7, 2025/Notices\n\nThe EA will be issued and circulated\nfor review by all interested parties. All\ncomments filed on the EA will be\nanalyzed by staff and considered in the\nCommission\u2019s final licensing decision.", "job_id": null}, "started_at": "2026-10-31T12:47:47.703714", "completed_at": "2026-10-31T12:47:47.703714", "latency_in_ms": 106048}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__1815.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_multicolumns__1815", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__1815.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "</s><s><predict_bbox><predict_classes><output_markdown><predict_text_in_pic><table><tr><th>WATERLOO</th>\nIN<th>HISTORIOGRAPHY</th></tr><tr><td><td rowspan=\"2\" rowspan=\"2\"; region=12000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000</s>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "</s><s><predict_bbox><predict_classes><output_markdown><predict_text_in_pic><table><tr><th>WATERLOO</th>\nIN<th>HISTORIOGRAPHY</th></tr><tr><td><td rowspan=\"2\" rowspan=\"2\"; region=12000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000</s>"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__1815", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "</s><s><predict_bbox><predict_classes><output_markdown><predict_text_in_pic><table><tr><th>WATERLOO</th>\nIN<th>HISTORIOGRAPHY</th></tr><tr><td><td rowspan=\"2\" rowspan=\"2\"; region=12000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000</s>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "</s><s><predict_bbox><predict_classes><output_markdown><predict_text_in_pic><table><tr><th>WATERLOO</th>\nIN<th>HISTORIOGRAPHY</th></tr><tr><td><td rowspan=\"2\" rowspan=\"2\"; region=12000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000</s>", "job_id": null}, "started_at": "2026-10-31T03:45:29.516253", "completed_at": "2026-10-31T03:45:29.516253", "latency_in_ms": 106110}

+ {"request": {"example_id": "text/text_multicolumns__1815", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__1815.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "</s><s><predict_bbox><predict_classes><output_markdown><predict_text_in_pic><table><tr><th>WATERLOO</th>\nIN<th>HISTORIOGRAPHY</th></tr><tr><td><td rowspan=\"2\" rowspan=\"2\"; region=12000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000</s>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "</s><s><predict_bbox><predict_classes><output_markdown><predict_text_in_pic><table><tr><th>WATERLOO</th>\nIN<th>HISTORIOGRAPHY</th></tr><tr><td><td rowspan=\"2\" rowspan=\"2\"; region=12000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000</s>"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__1815", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "</s><s><predict_bbox><predict_classes><output_markdown><predict_text_in_pic><table><tr><th>WATERLOO</th>\nIN<th>HISTORIOGRAPHY</th></tr><tr><td><td rowspan=\"2\" rowspan=\"2\"; region=12000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000</s>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "</s><s><predict_bbox><predict_classes><output_markdown><predict_text_in_pic><table><tr><th>WATERLOO</th>\nIN<th>HISTORIOGRAPHY</th></tr><tr><td><td rowspan=\"2\" rowspan=\"2\"; region=12000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000000</s>", "job_id": null}, "started_at": "2026-10-31T12:47:48.106439", "completed_at": "2026-10-31T12:47:48.106439", "latency_in_ms": 106110}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2cc.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_multicolumns__2cc", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__2cc.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "Bandorski D _et al._ Contraindications for capsule endoscopy\n\nimaging techniques before VCE in suspected Crohn\u2019s disease<sup>[22]</sup>. However, in 2015 based on broader evidence, the European Society for Gastrointestinal Endoscopy (ESGE) recommended not using cross sectional imaging or patency capsule before VCE in patients with suspected Crohn\u2019s disease in the absence of obstructive symptoms. In contrast, in established Crohn\u2019s disease, imaging techniques and patency capsule are recommended to precede VCE<sup>[23]</sup><sup>.</sup>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "Bandorski D _et al._ Contraindications for capsule endoscopy\n\nimaging techniques before VCE in suspected Crohn\u2019s disease<sup>[22]</sup>. However, in 2015 based on broader evidence, the European Society for Gastrointestinal Endoscopy (ESGE) recommended not using cross sectional imaging or patency capsule before VCE in patients with suspected Crohn\u2019s disease in the absence of obstructive symptoms. In contrast, in established Crohn\u2019s disease, imaging techniques and patency capsule are recommended to precede VCE<sup>[23]</sup><sup>.</sup>"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__2cc", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "Bandorski D _et al._ Contraindications for capsule endoscopy\n\nimaging techniques before VCE in suspected Crohn\u2019s disease<sup>[22]</sup>. However, in 2015 based on broader evidence, the European Society for Gastrointestinal Endoscopy (ESGE) recommended not using cross sectional imaging or patency capsule before VCE in patients with suspected Crohn\u2019s disease in the absence of obstructive symptoms. In contrast, in established Crohn\u2019s disease, imaging techniques and patency capsule are recommended to precede VCE<sup>[23]</sup><sup>.</sup>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "Bandorski D _et al._ Contraindications for capsule endoscopy\n\nimaging techniques before VCE in suspected Crohn\u2019s disease<sup>[22]</sup>. However, in 2015 based on broader evidence, the European Society for Gastrointestinal Endoscopy (ESGE) recommended not using cross sectional imaging or patency capsule before VCE in patients with suspected Crohn\u2019s disease in the absence of obstructive symptoms. In contrast, in established Crohn\u2019s disease, imaging techniques and patency capsule are recommended to precede VCE<sup>[23]</sup><sup>.</sup>", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:29.~~912891~~", "completed_at": "2026-10-~~31T03~~:45:29.~~912891~~", "latency_in_ms": 35335}

+ {"request": {"example_id": "text/text_multicolumns__2cc", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__2cc.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "Bandorski D _et al._ Contraindications for capsule endoscopy\n\nimaging techniques before VCE in suspected Crohn\u2019s disease<sup>[22]</sup>. However, in 2015 based on broader evidence, the European Society for Gastrointestinal Endoscopy (ESGE) recommended not using cross sectional imaging or patency capsule before VCE in patients with suspected Crohn\u2019s disease in the absence of obstructive symptoms. In contrast, in established Crohn\u2019s disease, imaging techniques and patency capsule are recommended to precede VCE<sup>[23]</sup><sup>.</sup>", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "Bandorski D _et al._ Contraindications for capsule endoscopy\n\nimaging techniques before VCE in suspected Crohn\u2019s disease<sup>[22]</sup>. However, in 2015 based on broader evidence, the European Society for Gastrointestinal Endoscopy (ESGE) recommended not using cross sectional imaging or patency capsule before VCE in patients with suspected Crohn\u2019s disease in the absence of obstructive symptoms. In contrast, in established Crohn\u2019s disease, imaging techniques and patency capsule are recommended to precede VCE<sup>[23]</sup><sup>.</sup>"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__2cc", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "Bandorski D _et al._ Contraindications for capsule endoscopy\n\nimaging techniques before VCE in suspected Crohn\u2019s disease<sup>[22]</sup>. However, in 2015 based on broader evidence, the European Society for Gastrointestinal Endoscopy (ESGE) recommended not using cross sectional imaging or patency capsule before VCE in patients with suspected Crohn\u2019s disease in the absence of obstructive symptoms. In contrast, in established Crohn\u2019s disease, imaging techniques and patency capsule are recommended to precede VCE<sup>[23]</sup><sup>.</sup>", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "Bandorski D _et al._ Contraindications for capsule endoscopy\n\nimaging techniques before VCE in suspected Crohn\u2019s disease<sup>[22]</sup>. However, in 2015 based on broader evidence, the European Society for Gastrointestinal Endoscopy (ESGE) recommended not using cross sectional imaging or patency capsule before VCE in patients with suspected Crohn\u2019s disease in the absence of obstructive symptoms. In contrast, in established Crohn\u2019s disease, imaging techniques and patency capsule are recommended to precede VCE<sup>[23]</sup><sup>.</sup>", "job_id": null}, "started_at": "2026-10-31T12:47:48.498434", "completed_at": "2026-10-31T12:47:48.498434", "latency_in_ms": 35335}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2col.result.json CHANGED Viewed

The diff for this file is too large to render. See raw diff

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2col_fund_summary.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_multicolumns__2col_fund_summary", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__2col_fund_summary.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# FSR<br>\nFlorida Retirement System", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# FSR<br>\nFlorida Retirement System"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__2col_fund_summary", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# FSR<br>\nFlorida Retirement System", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# FSR<br>\nFlorida Retirement System", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:30.~~609891~~", "completed_at": "2026-10-~~31T03~~:45:30.~~609891~~", "latency_in_ms": 105955}

+ {"request": {"example_id": "text/text_multicolumns__2col_fund_summary", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__2col_fund_summary.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# FSR<br>\nFlorida Retirement System", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# FSR<br>\nFlorida Retirement System"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__2col_fund_summary", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# FSR<br>\nFlorida Retirement System", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# FSR<br>\nFlorida Retirement System", "job_id": null}, "started_at": "2026-10-31T12:47:49.192293", "completed_at": "2026-10-31T12:47:49.192293", "latency_in_ms": 105955}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2col_paper.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_multicolumns__2col_paper", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__2col_paper.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Mind Your Tone: Investigating\nHow Prompt Politeness Affects\nLLM Accuracy (short paper)\n\nOm Dobariya<sup>1</sup> and **Akhil Kumar**<sup>1</sup>**<br> {okd5069@psu.edu,<br> akhilkumar@psu.edu}**\n\nAbstract<br> The wording of natural language prompts has been shown to influence the performance of large language models (LLMs), yet the role of politeness and tone remains underexplored. In this study, we investigate how varying levels of prompt politeness affect model accuracy on multiple-choice questions. We created a dataset of 50 base questions spanning mathematics, science, and history, each rewritten into five tone variants\u2014Very Polite, Polite, Neutral, Rude, and Very Rude\u2014yielding 250 unique prompts. Using ChatGPT-4o, we evaluated responses across these conditions and applied paired sample t-tests to assess statistical significance. Contrary to expectations, impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts. These findings differ from earlier studies that associated rudeness with poorer outcomes, suggesting that newer LLMs may respond differently to tonal variation. Our results highlight the importance of studying pragmatic aspects of prompting and raise broader questions about the social dimensions of human\u2013AI interaction.\n\nSince these powerful LLMs are accessed through a natural language interface, there are also several notions of how minor differences in inputs, formally called \u2018prompts\u2019, affect their response quality, as measured by accuracy, length, coherence, etc. Thus, a new field of study called \u2018prompt engineering\u2019 has emerged to study the variance in response quality from different prompt designs and create better prompts for the most desired results (Sclar et al., 2024).\n\nWork has been done in prompt engineering in recent years to study the effect of prompt structure, style, language, and other factors on the quality of the results (Yang, et al., 2023). One such factor is politeness in the wording of the prompt. Prior work demonstrates that varying levels of prompt politeness can cause statistically significant shifts in model accuracy across multiple languages and tasks (Yin et al., 2024). In this study, we revisit this notion to validate if politeness is a factor affecting the accuracy of LLMs using a dataset of 50 base questions with four options, one of which is correct. Each base question has five variants depending on the politeness level, \u2018Very Polite\u2019, \u2018Polite\u2019, \u2018Neutral\u2019, \u2018Rude\u2019, and \u2018Very Rude\u2019, making a total of 250 questions. We fed these 250 questions into ChatGPT 4o to examine the variance across politeness echelons.\n\nThis paper is organized as follows. Section 2 provides background and related work. Section 3 describes the dataset construction and explains our research methodology. Section 4 presents the results of our experiments. Section 5 offers discussion and conclusions. Section 6 outlines the limitations of our study, and Section 7 addresses the ethical considerations arising from our findings.\n\n## Background and Related Work\n\nSince OpenAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nBecause openAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nNaturally, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\nnecessarily, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\n<table>1</sub>Pennsylvania State University, University Park", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Mind Your Tone: Investigating\nHow Prompt Politeness Affects\nLLM Accuracy (short paper)\n\nOm Dobariya<sup>1</sup> and **Akhil Kumar**<sup>1</sup>**<br> {okd5069@psu.edu,<br> akhilkumar@psu.edu}**\n\nAbstract<br> The wording of natural language prompts has been shown to influence the performance of large language models (LLMs), yet the role of politeness and tone remains underexplored. In this study, we investigate how varying levels of prompt politeness affect model accuracy on multiple-choice questions. We created a dataset of 50 base questions spanning mathematics, science, and history, each rewritten into five tone variants\u2014Very Polite, Polite, Neutral, Rude, and Very Rude\u2014yielding 250 unique prompts. Using ChatGPT-4o, we evaluated responses across these conditions and applied paired sample t-tests to assess statistical significance. Contrary to expectations, impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts. These findings differ from earlier studies that associated rudeness with poorer outcomes, suggesting that newer LLMs may respond differently to tonal variation. Our results highlight the importance of studying pragmatic aspects of prompting and raise broader questions about the social dimensions of human\u2013AI interaction.\n\nSince these powerful LLMs are accessed through a natural language interface, there are also several notions of how minor differences in inputs, formally called \u2018prompts\u2019, affect their response quality, as measured by accuracy, length, coherence, etc. Thus, a new field of study called \u2018prompt engineering\u2019 has emerged to study the variance in response quality from different prompt designs and create better prompts for the most desired results (Sclar et al., 2024).\n\nWork has been done in prompt engineering in recent years to study the effect of prompt structure, style, language, and other factors on the quality of the results (Yang, et al., 2023). One such factor is politeness in the wording of the prompt. Prior work demonstrates that varying levels of prompt politeness can cause statistically significant shifts in model accuracy across multiple languages and tasks (Yin et al., 2024). In this study, we revisit this notion to validate if politeness is a factor affecting the accuracy of LLMs using a dataset of 50 base questions with four options, one of which is correct. Each base question has five variants depending on the politeness level, \u2018Very Polite\u2019, \u2018Polite\u2019, \u2018Neutral\u2019, \u2018Rude\u2019, and \u2018Very Rude\u2019, making a total of 250 questions. We fed these 250 questions into ChatGPT 4o to examine the variance across politeness echelons.\n\nThis paper is organized as follows. Section 2 provides background and related work. Section 3 describes the dataset construction and explains our research methodology. Section 4 presents the results of our experiments. Section 5 offers discussion and conclusions. Section 6 outlines the limitations of our study, and Section 7 addresses the ethical considerations arising from our findings.\n\n## Background and Related Work\n\nSince OpenAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nBecause openAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nNaturally, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\nnecessarily, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\n<table>1</sub>Pennsylvania State University, University Park"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__2col_paper", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Mind Your Tone: Investigating\nHow Prompt Politeness Affects\nLLM Accuracy (short paper)\n\nOm Dobariya<sup>1</sup> and **Akhil Kumar**<sup>1</sup>**<br> {okd5069@psu.edu,<br> akhilkumar@psu.edu}**\n\nAbstract<br> The wording of natural language prompts has been shown to influence the performance of large language models (LLMs), yet the role of politeness and tone remains underexplored. In this study, we investigate how varying levels of prompt politeness affect model accuracy on multiple-choice questions. We created a dataset of 50 base questions spanning mathematics, science, and history, each rewritten into five tone variants\u2014Very Polite, Polite, Neutral, Rude, and Very Rude\u2014yielding 250 unique prompts. Using ChatGPT-4o, we evaluated responses across these conditions and applied paired sample t-tests to assess statistical significance. Contrary to expectations, impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts. These findings differ from earlier studies that associated rudeness with poorer outcomes, suggesting that newer LLMs may respond differently to tonal variation. Our results highlight the importance of studying pragmatic aspects of prompting and raise broader questions about the social dimensions of human\u2013AI interaction.\n\nSince these powerful LLMs are accessed through a natural language interface, there are also several notions of how minor differences in inputs, formally called \u2018prompts\u2019, affect their response quality, as measured by accuracy, length, coherence, etc. Thus, a new field of study called \u2018prompt engineering\u2019 has emerged to study the variance in response quality from different prompt designs and create better prompts for the most desired results (Sclar et al., 2024).\n\nWork has been done in prompt engineering in recent years to study the effect of prompt structure, style, language, and other factors on the quality of the results (Yang, et al., 2023). One such factor is politeness in the wording of the prompt. Prior work demonstrates that varying levels of prompt politeness can cause statistically significant shifts in model accuracy across multiple languages and tasks (Yin et al., 2024). In this study, we revisit this notion to validate if politeness is a factor affecting the accuracy of LLMs using a dataset of 50 base questions with four options, one of which is correct. Each base question has five variants depending on the politeness level, \u2018Very Polite\u2019, \u2018Polite\u2019, \u2018Neutral\u2019, \u2018Rude\u2019, and \u2018Very Rude\u2019, making a total of 250 questions. We fed these 250 questions into ChatGPT 4o to examine the variance across politeness echelons.\n\nThis paper is organized as follows. Section 2 provides background and related work. Section 3 describes the dataset construction and explains our research methodology. Section 4 presents the results of our experiments. Section 5 offers discussion and conclusions. Section 6 outlines the limitations of our study, and Section 7 addresses the ethical considerations arising from our findings.\n\n## Background and Related Work\n\nSince OpenAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nBecause openAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nNaturally, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\nnecessarily, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\n<table>1</sub>Pennsylvania State University, University Park", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Mind Your Tone: Investigating\nHow Prompt Politeness Affects\nLLM Accuracy (short paper)\n\nOm Dobariya<sup>1</sup> and **Akhil Kumar**<sup>1</sup>**<br> {okd5069@psu.edu,<br> akhilkumar@psu.edu}**\n\nAbstract<br> The wording of natural language prompts has been shown to influence the performance of large language models (LLMs), yet the role of politeness and tone remains underexplored. In this study, we investigate how varying levels of prompt politeness affect model accuracy on multiple-choice questions. We created a dataset of 50 base questions spanning mathematics, science, and history, each rewritten into five tone variants\u2014Very Polite, Polite, Neutral, Rude, and Very Rude\u2014yielding 250 unique prompts. Using ChatGPT-4o, we evaluated responses across these conditions and applied paired sample t-tests to assess statistical significance. Contrary to expectations, impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts. These findings differ from earlier studies that associated rudeness with poorer outcomes, suggesting that newer LLMs may respond differently to tonal variation. Our results highlight the importance of studying pragmatic aspects of prompting and raise broader questions about the social dimensions of human\u2013AI interaction.\n\nSince these powerful LLMs are accessed through a natural language interface, there are also several notions of how minor differences in inputs, formally called \u2018prompts\u2019, affect their response quality, as measured by accuracy, length, coherence, etc. Thus, a new field of study called \u2018prompt engineering\u2019 has emerged to study the variance in response quality from different prompt designs and create better prompts for the most desired results (Sclar et al., 2024).\n\nWork has been done in prompt engineering in recent years to study the effect of prompt structure, style, language, and other factors on the quality of the results (Yang, et al., 2023). One such factor is politeness in the wording of the prompt. Prior work demonstrates that varying levels of prompt politeness can cause statistically significant shifts in model accuracy across multiple languages and tasks (Yin et al., 2024). In this study, we revisit this notion to validate if politeness is a factor affecting the accuracy of LLMs using a dataset of 50 base questions with four options, one of which is correct. Each base question has five variants depending on the politeness level, \u2018Very Polite\u2019, \u2018Polite\u2019, \u2018Neutral\u2019, \u2018Rude\u2019, and \u2018Very Rude\u2019, making a total of 250 questions. We fed these 250 questions into ChatGPT 4o to examine the variance across politeness echelons.\n\nThis paper is organized as follows. Section 2 provides background and related work. Section 3 describes the dataset construction and explains our research methodology. Section 4 presents the results of our experiments. Section 5 offers discussion and conclusions. Section 6 outlines the limitations of our study, and Section 7 addresses the ethical considerations arising from our findings.\n\n## Background and Related Work\n\nSince OpenAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nBecause openAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nNaturally, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\nnecessarily, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\n<table>1</sub>Pennsylvania State University, University Park", "job_id": null}, "started_at": "2026-10-31T03:45:30.947317", "completed_at": "2026-10-31T03:45:30.947317", "latency_in_ms": 11978}

+ {"request": {"example_id": "text/text_multicolumns__2col_paper", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__2col_paper.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Mind Your Tone: Investigating\nHow Prompt Politeness Affects\nLLM Accuracy (short paper)\n\nOm Dobariya<sup>1</sup> and **Akhil Kumar**<sup>1</sup>**<br> {okd5069@psu.edu,<br> akhilkumar@psu.edu}**\n\nAbstract<br> The wording of natural language prompts has been shown to influence the performance of large language models (LLMs), yet the role of politeness and tone remains underexplored. In this study, we investigate how varying levels of prompt politeness affect model accuracy on multiple-choice questions. We created a dataset of 50 base questions spanning mathematics, science, and history, each rewritten into five tone variants\u2014Very Polite, Polite, Neutral, Rude, and Very Rude\u2014yielding 250 unique prompts. Using ChatGPT-4o, we evaluated responses across these conditions and applied paired sample t-tests to assess statistical significance. Contrary to expectations, impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts. These findings differ from earlier studies that associated rudeness with poorer outcomes, suggesting that newer LLMs may respond differently to tonal variation. Our results highlight the importance of studying pragmatic aspects of prompting and raise broader questions about the social dimensions of human\u2013AI interaction.\n\nSince these powerful LLMs are accessed through a natural language interface, there are also several notions of how minor differences in inputs, formally called \u2018prompts\u2019, affect their response quality, as measured by accuracy, length, coherence, etc. Thus, a new field of study called \u2018prompt engineering\u2019 has emerged to study the variance in response quality from different prompt designs and create better prompts for the most desired results (Sclar et al., 2024).\n\nWork has been done in prompt engineering in recent years to study the effect of prompt structure, style, language, and other factors on the quality of the results (Yang, et al., 2023). One such factor is politeness in the wording of the prompt. Prior work demonstrates that varying levels of prompt politeness can cause statistically significant shifts in model accuracy across multiple languages and tasks (Yin et al., 2024). In this study, we revisit this notion to validate if politeness is a factor affecting the accuracy of LLMs using a dataset of 50 base questions with four options, one of which is correct. Each base question has five variants depending on the politeness level, \u2018Very Polite\u2019, \u2018Polite\u2019, \u2018Neutral\u2019, \u2018Rude\u2019, and \u2018Very Rude\u2019, making a total of 250 questions. We fed these 250 questions into ChatGPT 4o to examine the variance across politeness echelons.\n\nThis paper is organized as follows. Section 2 provides background and related work. Section 3 describes the dataset construction and explains our research methodology. Section 4 presents the results of our experiments. Section 5 offers discussion and conclusions. Section 6 outlines the limitations of our study, and Section 7 addresses the ethical considerations arising from our findings.\n\n## Background and Related Work\n\nSince OpenAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nBecause openAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nNaturally, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\nnecessarily, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\n<table>1</sub>Pennsylvania State University, University Park", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Mind Your Tone: Investigating\nHow Prompt Politeness Affects\nLLM Accuracy (short paper)\n\nOm Dobariya<sup>1</sup> and **Akhil Kumar**<sup>1</sup>**<br> {okd5069@psu.edu,<br> akhilkumar@psu.edu}**\n\nAbstract<br> The wording of natural language prompts has been shown to influence the performance of large language models (LLMs), yet the role of politeness and tone remains underexplored. In this study, we investigate how varying levels of prompt politeness affect model accuracy on multiple-choice questions. We created a dataset of 50 base questions spanning mathematics, science, and history, each rewritten into five tone variants\u2014Very Polite, Polite, Neutral, Rude, and Very Rude\u2014yielding 250 unique prompts. Using ChatGPT-4o, we evaluated responses across these conditions and applied paired sample t-tests to assess statistical significance. Contrary to expectations, impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts. These findings differ from earlier studies that associated rudeness with poorer outcomes, suggesting that newer LLMs may respond differently to tonal variation. Our results highlight the importance of studying pragmatic aspects of prompting and raise broader questions about the social dimensions of human\u2013AI interaction.\n\nSince these powerful LLMs are accessed through a natural language interface, there are also several notions of how minor differences in inputs, formally called \u2018prompts\u2019, affect their response quality, as measured by accuracy, length, coherence, etc. Thus, a new field of study called \u2018prompt engineering\u2019 has emerged to study the variance in response quality from different prompt designs and create better prompts for the most desired results (Sclar et al., 2024).\n\nWork has been done in prompt engineering in recent years to study the effect of prompt structure, style, language, and other factors on the quality of the results (Yang, et al., 2023). One such factor is politeness in the wording of the prompt. Prior work demonstrates that varying levels of prompt politeness can cause statistically significant shifts in model accuracy across multiple languages and tasks (Yin et al., 2024). In this study, we revisit this notion to validate if politeness is a factor affecting the accuracy of LLMs using a dataset of 50 base questions with four options, one of which is correct. Each base question has five variants depending on the politeness level, \u2018Very Polite\u2019, \u2018Polite\u2019, \u2018Neutral\u2019, \u2018Rude\u2019, and \u2018Very Rude\u2019, making a total of 250 questions. We fed these 250 questions into ChatGPT 4o to examine the variance across politeness echelons.\n\nThis paper is organized as follows. Section 2 provides background and related work. Section 3 describes the dataset construction and explains our research methodology. Section 4 presents the results of our experiments. Section 5 offers discussion and conclusions. Section 6 outlines the limitations of our study, and Section 7 addresses the ethical considerations arising from our findings.\n\n## Background and Related Work\n\nSince OpenAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nBecause openAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nNaturally, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\nnecessarily, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\n<table>1</sub>Pennsylvania State University, University Park"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__2col_paper", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Mind Your Tone: Investigating\nHow Prompt Politeness Affects\nLLM Accuracy (short paper)\n\nOm Dobariya<sup>1</sup> and **Akhil Kumar**<sup>1</sup>**<br> {okd5069@psu.edu,<br> akhilkumar@psu.edu}**\n\nAbstract<br> The wording of natural language prompts has been shown to influence the performance of large language models (LLMs), yet the role of politeness and tone remains underexplored. In this study, we investigate how varying levels of prompt politeness affect model accuracy on multiple-choice questions. We created a dataset of 50 base questions spanning mathematics, science, and history, each rewritten into five tone variants\u2014Very Polite, Polite, Neutral, Rude, and Very Rude\u2014yielding 250 unique prompts. Using ChatGPT-4o, we evaluated responses across these conditions and applied paired sample t-tests to assess statistical significance. Contrary to expectations, impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts. These findings differ from earlier studies that associated rudeness with poorer outcomes, suggesting that newer LLMs may respond differently to tonal variation. Our results highlight the importance of studying pragmatic aspects of prompting and raise broader questions about the social dimensions of human\u2013AI interaction.\n\nSince these powerful LLMs are accessed through a natural language interface, there are also several notions of how minor differences in inputs, formally called \u2018prompts\u2019, affect their response quality, as measured by accuracy, length, coherence, etc. Thus, a new field of study called \u2018prompt engineering\u2019 has emerged to study the variance in response quality from different prompt designs and create better prompts for the most desired results (Sclar et al., 2024).\n\nWork has been done in prompt engineering in recent years to study the effect of prompt structure, style, language, and other factors on the quality of the results (Yang, et al., 2023). One such factor is politeness in the wording of the prompt. Prior work demonstrates that varying levels of prompt politeness can cause statistically significant shifts in model accuracy across multiple languages and tasks (Yin et al., 2024). In this study, we revisit this notion to validate if politeness is a factor affecting the accuracy of LLMs using a dataset of 50 base questions with four options, one of which is correct. Each base question has five variants depending on the politeness level, \u2018Very Polite\u2019, \u2018Polite\u2019, \u2018Neutral\u2019, \u2018Rude\u2019, and \u2018Very Rude\u2019, making a total of 250 questions. We fed these 250 questions into ChatGPT 4o to examine the variance across politeness echelons.\n\nThis paper is organized as follows. Section 2 provides background and related work. Section 3 describes the dataset construction and explains our research methodology. Section 4 presents the results of our experiments. Section 5 offers discussion and conclusions. Section 6 outlines the limitations of our study, and Section 7 addresses the ethical considerations arising from our findings.\n\n## Background and Related Work\n\nSince OpenAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nBecause openAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nNaturally, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\nnecessarily, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\n<table>1</sub>Pennsylvania State University, University Park", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Mind Your Tone: Investigating\nHow Prompt Politeness Affects\nLLM Accuracy (short paper)\n\nOm Dobariya<sup>1</sup> and **Akhil Kumar**<sup>1</sup>**<br> {okd5069@psu.edu,<br> akhilkumar@psu.edu}**\n\nAbstract<br> The wording of natural language prompts has been shown to influence the performance of large language models (LLMs), yet the role of politeness and tone remains underexplored. In this study, we investigate how varying levels of prompt politeness affect model accuracy on multiple-choice questions. We created a dataset of 50 base questions spanning mathematics, science, and history, each rewritten into five tone variants\u2014Very Polite, Polite, Neutral, Rude, and Very Rude\u2014yielding 250 unique prompts. Using ChatGPT-4o, we evaluated responses across these conditions and applied paired sample t-tests to assess statistical significance. Contrary to expectations, impolite prompts consistently outperformed polite ones, with accuracy ranging from 80.8% for Very Polite prompts to 84.8% for Very Rude prompts. These findings differ from earlier studies that associated rudeness with poorer outcomes, suggesting that newer LLMs may respond differently to tonal variation. Our results highlight the importance of studying pragmatic aspects of prompting and raise broader questions about the social dimensions of human\u2013AI interaction.\n\nSince these powerful LLMs are accessed through a natural language interface, there are also several notions of how minor differences in inputs, formally called \u2018prompts\u2019, affect their response quality, as measured by accuracy, length, coherence, etc. Thus, a new field of study called \u2018prompt engineering\u2019 has emerged to study the variance in response quality from different prompt designs and create better prompts for the most desired results (Sclar et al., 2024).\n\nWork has been done in prompt engineering in recent years to study the effect of prompt structure, style, language, and other factors on the quality of the results (Yang, et al., 2023). One such factor is politeness in the wording of the prompt. Prior work demonstrates that varying levels of prompt politeness can cause statistically significant shifts in model accuracy across multiple languages and tasks (Yin et al., 2024). In this study, we revisit this notion to validate if politeness is a factor affecting the accuracy of LLMs using a dataset of 50 base questions with four options, one of which is correct. Each base question has five variants depending on the politeness level, \u2018Very Polite\u2019, \u2018Polite\u2019, \u2018Neutral\u2019, \u2018Rude\u2019, and \u2018Very Rude\u2019, making a total of 250 questions. We fed these 250 questions into ChatGPT 4o to examine the variance across politeness echelons.\n\nThis paper is organized as follows. Section 2 provides background and related work. Section 3 describes the dataset construction and explains our research methodology. Section 4 presents the results of our experiments. Section 5 offers discussion and conclusions. Section 6 outlines the limitations of our study, and Section 7 addresses the ethical considerations arising from our findings.\n\n## Background and Related Work\n\nSince OpenAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nBecause openAI launched its LLM ChatGPT-3.5 in November 2022, Artificial Intelligence (AI) has become a household name, given the unprecedented abilities at tackling demanding tasks. LLMs are models that typically take in text-based prompts and output text; however, with the recent advancements, LLMs can now handle more than text data modality, which gave them a new name - multi-modal models (Huyen, 2024).\n\nNaturally, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\nnecessarily, with any new great invention comes the question of maximizing the output result. The same is true for LLMs with the idea of prompt engineering \u2013 the science of developing various\n\n<table>1</sub>Pennsylvania State University, University Park", "job_id": null}, "started_at": "2026-10-31T12:47:49.531032", "completed_at": "2026-10-31T12:47:49.531032", "latency_in_ms": 11978}

benchmarks/v1/parsebench_results/thrisul-parser/text/text_text_multicolumns__2col_repport.result.json CHANGED Viewed

@@ -1 +1 @@

- {"request": {"example_id": "text/text_multicolumns__2col_repport", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__2col_repport.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Bridgepoint \u2013 2022 Annual Report & Accounts\n\n# Chairman\u2019s statement\nWilliam Jackson<x_0.5742>\n\nWilliam Jackson", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Bridgepoint \u2013 2022 Annual Report & Accounts\n\n# Chairman\u2019s statement\nWilliam Jackson<x_0.5742>\n\nWilliam Jackson"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__2col_repport", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Bridgepoint \u2013 2022 Annual Report & Accounts\n\n# Chairman\u2019s statement\nWilliam Jackson<x_0.5742>\n\nWilliam Jackson", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Bridgepoint \u2013 2022 Annual Report & Accounts\n\n# Chairman\u2019s statement\nWilliam Jackson<x_0.5742>\n\nWilliam Jackson", "job_id": null}, "started_at": "2026-10-~~31T03~~:45:31.~~328984~~", "completed_at": "2026-10-~~31T03~~:45:31.~~328984~~", "latency_in_ms": 106039}

+ {"request": {"example_id": "text/text_multicolumns__2col_repport", "source_file_path": "/kaggle/tmp/parsebench/docs/text/text_multicolumns__2col_repport.pdf", "product_type": "parse", "schema_override": null, "config_override": null}, "pipeline_name": "thrisul-parser", "product_type": "parse", "raw_output": {"markdown": "# Bridgepoint \u2013 2022 Annual Report & Accounts\n\n# Chairman\u2019s statement\nWilliam Jackson<x_0.5742>\n\nWilliam Jackson", "provider": "thrisul_bench_runner", "system": "thr", "raw_text": "# Bridgepoint \u2013 2022 Annual Report & Accounts\n\n# Chairman\u2019s statement\nWilliam Jackson<x_0.5742>\n\nWilliam Jackson"}, "output": {"task_type": "parse", "example_id": "text/text_multicolumns__2col_repport", "pipeline_name": "thrisul-parser", "pages": [{"page_index": 0, "markdown": "# Bridgepoint \u2013 2022 Annual Report & Accounts\n\n# Chairman\u2019s statement\nWilliam Jackson<x_0.5742>\n\nWilliam Jackson", "line_numbers": []}], "layout_pages": [], "grounded_pages": [], "markdown": "# Bridgepoint \u2013 2022 Annual Report & Accounts\n\n# Chairman\u2019s statement\nWilliam Jackson<x_0.5742>\n\nWilliam Jackson", "job_id": null}, "started_at": "2026-10-31T12:47:49.924342", "completed_at": "2026-10-31T12:47:49.924342", "latency_in_ms": 106039}

benchmarks/v1/results.json CHANGED Viewed

@@ -285,44 +285,285 @@
   "olmbench": {
     "scores": {
       "nemotron-base": {
-        "Overall": 0.0,
-        "scorer_status": "FAILED (errors)",
-        "headers_footers": 0.0,
-        "long_tiny_text": 0.0,
-        "multi_column": 0.0,
-        "old_scans": 0.0,
-        "table_tests": 0.0
       },
       "thrisul-parser": {
-        "Overall": 0.0,
-        "scorer_status": "FAILED (errors)",
-        "headers_footers": 0.0,
-        "long_tiny_text": 0.0,
-        "multi_column": 0.0,
-        "old_scans": 0.0,
-        "table_tests": 0.0
       }
     },
     "pages_covered": 400,
     "pages_total": 1403,
     "repeats": 1,
-    "math_categories_pending": [
-      "arxiv_math",
-      "old_scans_math"
-    ],
-    "note": "official scorer on the covered page subset (filtered bench dir, category-nested like the official layout; only covered tests included so partial coverage is never counted as failure); full-bench run when OLM_PAGES covers all pages; math categories moved aside (no usable browser) and recorded as pending"
   },
   "omnidocbench": {
     "status": "packaged",
-    "pages_covered": 82,
     "pages_total": 1651,
     "mode": "package"
   },
   "nvidia_suites": {
     "IndicVisionBench": {
-      "repo": "nvidia/IndicVisionBench",
-      "status": "external_pending",
-      "reason": "RuntimeError('no loadable split')"
     },
     "MOSCAR": {
       "repo": "nvidia/MOSCAR",
@@ -340,14 +581,20 @@
     {
       "suite": "olmOCR-Bench (official scorer)",
       "metric": "Overall pass rate",
-      "value": 0.0,
-      "detail": "scorer_status=FAILED (errors), headers_footers=0.0, long_tiny_text=0.0, multi_column=0.0, old_scans=0.0, table_tests=0.0"
     },
     {
       "suite": "OmniDocBench",
       "metric": "status",
       "value": "packaged",
-      "detail": "82 pages packaged"
     }
   ],
   "config_public": {
@@ -365,12 +612,12 @@
     "OMNI_MODE": "package",
     "OMNI_PAGES": 120,
     "NVIDIA_SUITES": true,
-    "INDIC_PAGES": 200,
     "MOSCAR_PAGES": 200,
     "SESSION_HOURS": 11,
     "SAFETY_MIN": 30,
     "SEED": 1729,
-    "OLM_TIME_SHARE": 0.55,
     "OMNI_TIME_SHARE": 0.15,
     "MAX_NEW_TOKENS": 8744,
     "GEN_BATCH": 4,

   "olmbench": {
     "scores": {
       "nemotron-base": {
+        "Overall": 55.4,
+        "CI95_half_width": 1.6,
+        "absent": 24.7,
+        "baseline": 99.8,
+        "math": 24.8,
+        "order": 62.5,
+        "present": 61.1,
+        "table": 85.8,
+        "arxiv_math": 42.6,
+        "headers_footers": 18.3,
+        "long_tiny_text": 69.4,
+        "multi_column": 79.6,
+        "old_scans": 36.5,
+        "old_scans_math": 10.9,
+        "table_tests": 85.8
       },
       "thrisul-parser": {
+        "Overall": 29.5,
+        "CI95_half_width": 1.5,
+        "absent": 61.2,
+        "baseline": 98.5,
+        "math": 4.9,
+        "order": 14.0,
+        "present": 15.1,
+        "table": 18.0,
+        "arxiv_math": 7.6,
+        "headers_footers": 56.7,
+        "long_tiny_text": 14.0,
+        "multi_column": 13.5,
+        "old_scans": 24.8,
+        "old_scans_math": 2.8,
+        "table_tests": 18.3
       }
     },
     "pages_covered": 400,
     "pages_total": 1403,
     "repeats": 1,
+    "math_categories_pending": [],
+    "scorer_error": null,
+    "note": "official scorer on the covered page subset (filtered bench dir, category-nested like the official layout; only covered tests included so partial coverage is never counted as failure); full-bench run when OLM_PAGES covers all pages"
   },
   "omnidocbench": {
     "status": "packaged",
+    "pages_covered": 119,
     "pages_total": 1651,
     "mode": "package"
   },
+  "indicvisionbench": {
+    "repo": "krutrim-ai-labs/IndicVisionBench",
+    "config": "ocr",
+    "rows_total": 876,
+    "agg": {
+      "thr": {
+        "n": 862,
+        "anls_c": 53.05217098342906,
+        "anls_w": 35.851067317190065,
+        "cer": 311.26781794770494,
+        "wer": 395.7415294037141,
+        "languages": {
+          "Bengali": {
+            "n": 87,
+            "anls_c": 72.06516758245996,
+            "cer": 444.95421128065567,
+            "wer": 684.6583685892634
+          },
+          "Gujarati": {
+            "n": 87,
+            "anls_c": 25.395088984596185,
+            "cer": 209.57549478261754,
+            "wer": 190.5224639648038
+          },
+          "Hindi": {
+            "n": 85,
+            "anls_c": 65.66241669091694,
+            "cer": 60.636138488309626,
+            "wer": 81.11222945285022
+          },
+          "Kannada": {
+            "n": 87,
+            "anls_c": 49.51308891812906,
+            "cer": 156.02995901010712,
+            "wer": 260.1300609459596
+          },
+          "Malayalam": {
+            "n": 87,
+            "anls_c": 24.05802589091149,
+            "cer": 726.8146220753227,
+            "wer": 1154.376255026419
+          },
+          "Marathi": {
+            "n": 83,
+            "anls_c": 65.32195640599434,
+            "cer": 156.37975349507633,
+            "wer": 113.64312571513373
+          },
+          "Odia": {
+            "n": 86,
+            "anls_c": 52.341048119421174,
+            "cer": 273.37765182025953,
+            "wer": 245.6566005538036
+          },
+          "Punjabi": {
+            "n": 87,
+            "anls_c": 52.858656510737276,
+            "cer": 521.8318568547596,
+            "wer": 419.7764872925679
+          },
+          "Tamil": {
+            "n": 87,
+            "anls_c": 62.188715995893915,
+            "cer": 254.4188333041377,
+            "wer": 321.4142716454671
+          },
+          "Telugu": {
+            "n": 86,
+            "anls_c": 62.06700850259098,
+            "cer": 295.1560078723313,
+            "wer": 464.99341640832927
+          }
+        }
+      },
+      "base": {
+        "n": 862,
+        "anls_c": 84.24201955114572,
+        "anls_w": 59.592097935780224,
+        "cer": 18.812691968530995,
+        "wer": 44.104446970786185,
+        "languages": {
+          "Bengali": {
+            "n": 87,
+            "anls_c": 91.29341684106654,
+            "cer": 8.734016927226289,
+            "wer": 27.11185512703557
+          },
+          "Gujarati": {
+            "n": 87,
+            "anls_c": 68.90393686572908,
+            "cer": 31.457572681386385,
+            "wer": 51.884469175777824
+          },
+          "Hindi": {
+            "n": 85,
+            "anls_c": 88.1959578537416,
+            "cer": 17.19627947789589,
+            "wer": 33.81500637522201
+          },
+          "Kannada": {
+            "n": 87,
+            "anls_c": 91.28174245217781,
+            "cer": 11.079595071694719,
+            "wer": 36.30255968108921
+          },
+          "Malayalam": {
+            "n": 87,
+            "anls_c": 63.26974511619004,
+            "cer": 46.95544767307656,
+            "wer": 88.60859663143759
+          },
+          "Marathi": {
+            "n": 83,
+            "anls_c": 93.78676515552064,
+            "cer": 6.2677187430514785,
+            "wer": 21.96948798789085
+          },
+          "Odia": {
+            "n": 86,
+            "anls_c": 73.76072875904623,
+            "cer": 29.935156676244137,
+            "wer": 65.36024031853114
+          },
+          "Punjabi": {
+            "n": 87,
+            "anls_c": 92.87697934201188,
+            "cer": 7.2951501051650265,
+            "wer": 20.650555500376115
+          },
+          "Tamil": {
+            "n": 87,
+            "anls_c": 91.60323439864274,
+            "cer": 12.556682263438667,
+            "wer": 37.857180997009245
+          },
+          "Telugu": {
+            "n": 86,
+            "anls_c": 87.89898218554887,
+            "cer": 16.132397216388053,
+            "wer": 56.61843822939625
+          }
+        }
+      }
+    }
+  },
   "nvidia_suites": {
     "IndicVisionBench": {
+      "repo": "krutrim-ai-labs/IndicVisionBench",
+      "config": "ocr",
+      "metric": "ANLS-C",
+      "n": 862,
+      "mean": 53.05217098342906,
+      "sec_per_page": 39.43873929596383,
+      "systems": {
+        "thr": {
+          "n": 862,
+          "anls_c": 53.05217098342906,
+          "anls_w": 35.851067317190065,
+          "cer": 311.26781794770494,
+          "wer": 395.7415294037141
+        },
+        "base": {
+          "n": 862,
+          "anls_c": 84.24201955114572,
+          "anls_w": 59.592097935780224,
+          "cer": 18.812691968530995,
+          "wer": 44.104446970786185
+        }
+      },
+      "languages": {
+        "Bengali": {
+          "n": 87,
+          "anls_c": 72.06516758245996,
+          "cer": 444.95421128065567,
+          "wer": 684.6583685892634
+        },
+        "Gujarati": {
+          "n": 87,
+          "anls_c": 25.395088984596185,
+          "cer": 209.57549478261754,
+          "wer": 190.5224639648038
+        },
+        "Hindi": {
+          "n": 85,
+          "anls_c": 65.66241669091694,
+          "cer": 60.636138488309626,
+          "wer": 81.11222945285022
+        },
+        "Kannada": {
+          "n": 87,
+          "anls_c": 49.51308891812906,
+          "cer": 156.02995901010712,
+          "wer": 260.1300609459596
+        },
+        "Malayalam": {
+          "n": 87,
+          "anls_c": 24.05802589091149,
+          "cer": 726.8146220753227,
+          "wer": 1154.376255026419
+        },
+        "Marathi": {
+          "n": 83,
+          "anls_c": 65.32195640599434,
+          "cer": 156.37975349507633,
+          "wer": 113.64312571513373
+        },
+        "Odia": {
+          "n": 86,
+          "anls_c": 52.341048119421174,
+          "cer": 273.37765182025953,
+          "wer": 245.6566005538036
+        },
+        "Punjabi": {
+          "n": 87,
+          "anls_c": 52.858656510737276,
+          "cer": 521.8318568547596,
+          "wer": 419.7764872925679
+        },
+        "Tamil": {
+          "n": 87,
+          "anls_c": 62.188715995893915,
+          "cer": 254.4188333041377,
+          "wer": 321.4142716454671
+        },
+        "Telugu": {
+          "n": 86,
+          "anls_c": 62.06700850259098,
+          "cer": 295.1560078723313,
+          "wer": 464.99341640832927
+        }
+      },
+      "detail": "per-language ANLS-C (thr): Bengali=72.1, Gujarati=25.4, Hindi=65.7, Kannada=49.5, Malayalam=24.1, Marathi=65.3, Odia=52.3, Punjabi=52.9, Tamil=62.2, Telugu=62.1 | overall CER 311.3, WER 395.7"
     },
     "MOSCAR": {
       "repo": "nvidia/MOSCAR",
     {
       "suite": "olmOCR-Bench (official scorer)",
       "metric": "Overall pass rate",
+      "value": 29.5,
+      "detail": "CI95_half_width=1.5, absent=61.2, baseline=98.5, math=4.9, order=14.0, present=15.1, table=18.0, arxiv_math=7.6, headers_footers=56.7, long_tiny_text=14.0, multi_column=13.5, old_scans=24.8, old_scans_math=2.8, table_tests=18.3"
     },
     {
       "suite": "OmniDocBench",
       "metric": "status",
       "value": "packaged",
+      "detail": "119 pages packaged"
+    },
+    {
+      "suite": "IndicVisionBench (our protocol)",
+      "metric": "ANLS-C",
+      "value": 53.05,
+      "detail": "per-language ANLS-C (thr): Bengali=72.1, Gujarati=25.4, Hindi=65.7, Kannada=49.5, Malayalam=24.1, Marathi=65.3, Odia=52.3, Punjabi=52.9, Tamil=62.2, Telugu=62.1 | overall CER 311.3, WER 395.7"
     }
   ],
   "config_public": {
     "OMNI_MODE": "package",
     "OMNI_PAGES": 120,
     "NVIDIA_SUITES": true,
+    "INDIC_PAGES": 876,
     "MOSCAR_PAGES": 200,
     "SESSION_HOURS": 11,
     "SAFETY_MIN": 30,
     "SEED": 1729,
+    "OLM_TIME_SHARE": 0.3,
     "OMNI_TIME_SHARE": 0.15,
     "MAX_NEW_TOKENS": 8744,
     "GEN_BATCH": 4,