Datasets:

KAUSTAcademy
/

RecipeMatching_eval-artifacts

ArXiv:

Tags:

retrieval

mathematics

embeddings

License:

Dataset card Data Studio Files Files and versions

xet

Community

Dataset Preview

Duplicate

Split (1)

train

The full dataset viewer is not available (click to read why). Only showing a preview of the rows.

The dataset generation failed

Error code:   DatasetGenerationError
Exception:    CastError
Message:      Couldn't cast
qid: string
gold_rank: int64
gold_sim: double
top10_ids: list<item: string>
  child 0, item: string
n_docs: int64
min_confidence: string
same_rank: int64
lang: string
overlaps_retrieve_anchor: bool
cluster_id: string
xling_rank: int64
to
{'qid': Value('string'), 'cluster_id': Value('string'), 'lang': Value('string'), 'min_confidence': Value('string'), 'overlaps_retrieve_anchor': Value('bool'), 'gold_rank': Value('int64'), 'xling_rank': Value('int64'), 'same_rank': Value('int64'), 'n_docs': Value('int64')}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1827, in _prepare_split_single
                  for key, table in generator:
                                    ^^^^^^^^^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 613, in wrapped
                  for item in generator(*args, **kwargs):
                              ~~~~~~~~~^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 343, in _generate_tables
                  self._cast_table(pa_table, json_field_paths=json_field_paths),
                  ~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/packaged_modules/json/json.py", line 132, in _cast_table
                  pa_table = table_cast(pa_table, self.info.features.arrow_schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2378, in table_cast
                  return cast_table_to_schema(table, schema)
                File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2306, in cast_table_to_schema
                  raise CastError(
                  ...<3 lines>...
                  )
              datasets.table.CastError: Couldn't cast
              qid: string
              gold_rank: int64
              gold_sim: double
              top10_ids: list<item: string>
                child 0, item: string
              n_docs: int64
              min_confidence: string
              same_rank: int64
              lang: string
              overlaps_retrieve_anchor: bool
              cluster_id: string
              xling_rank: int64
              to
              {'qid': Value('string'), 'cluster_id': Value('string'), 'lang': Value('string'), 'min_confidence': Value('string'), 'overlaps_retrieve_anchor': Value('bool'), 'gold_rank': Value('int64'), 'xling_rank': Value('int64'), 'same_rank': Value('int64'), 'n_docs': Value('int64')}
              because column names don't match
              
              The above exception was the direct cause of the following exception:
              
              Traceback (most recent call last):
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 1369, in compute_config_parquet_and_info_response
                  parquet_operations, partial, estimated_dataset_info = stream_convert_to_parquet(
                                                                        ~~~~~~~~~~~~~~~~~~~~~~~~~^
                      builder, max_dataset_size_bytes=max_dataset_size_bytes
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  )
                  ^
                File "/src/services/worker/src/worker/job_runners/config/parquet_and_info.py", line 948, in stream_convert_to_parquet
                  builder._prepare_split(split_generator=splits_generators[split], file_format="parquet")
                  ~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1694, in _prepare_split
                  for job_id, done, content in self._prepare_split_single(
                                               ~~~~~~~~~~~~~~~~~~~~~~~~~~^
                      gen_kwargs=gen_kwargs, job_id=job_id, **_prepare_split_args
                      ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
                  ):
                  ^
                File "/usr/local/lib/python3.14/site-packages/datasets/builder.py", line 1880, in _prepare_split_single
                  raise DatasetGenerationError("An error occurred while generating the dataset") from e
              datasets.exceptions.DatasetGenerationError: An error occurred while generating the dataset

Need help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.

qid string	cluster_id string	lang string	min_confidence string	overlaps_retrieve_anchor bool	xling_rank int64	n_docs int64
0bn0	0bn0	unknown	cas_verified	false	31,011	31,011
07t5	07t5	unknown	cas_verified	false	31,011	31,011
01p4	01p4	unknown	cas_verified	false	31,011	31,011
0gt0	0gt0	unknown	cas_verified	false	31,011	31,011
016x	016x	unknown	cas_verified	false	31,011	31,011
0c74	0c74	English	cas_verified	false	31,011	31,011
0dto	0dto	unknown	cas_verified	false	31,011	31,011
068c	068c	English	cas_verified	false	31,011	31,011
0cu4	0cu4	English; Russian	cas_verified	false	31,011	31,011
0584	0584	English	cas_verified	false	31,011	31,011
0bs1	0bs1	English	cas_verified	false	31,011	31,011
0bts	0bts	English	cas_verified	false	31,011	31,011
03fk	03fk	English	cas_verified	false	31,011	31,011
03r5	03r5	English	cas_verified	false	31,011	31,011
0gsh	0gsh	English	cas_verified	false	31,011	31,011
03a9	03a9	English	cas_verified	false	31,011	31,011
0hew	0hew	unknown	cas_verified	false	31,011	31,011
0lcq	0lcq	English	cas_verified	false	31,011	31,011
07oe	07oe	unknown	cas_verified	false	31,011	31,011
00p7	00p7	English	cas_verified	false	31,011	31,011
0168	0168	unknown	cas_verified	false	31,011	31,011
0e4a	0e4a	unknown	cas_verified	false	31,011	31,011
0655	0655	English	cas_verified	false	31,011	31,011
0h0m	0h0m	English	cas_verified	false	31,011	31,011
03bk	03bk	English	cas_verified	false	31,011	31,011
06n2	06n2	English	cas_verified	false	31,011	31,011
0cyi	0cyi	English	cas_verified	false	31,011	31,011
0byk	0byk	English	cas_verified	false	31,011	31,011
09xn	09xn	English	cas_verified	false	31,011	31,011
03xk	03xk	English	cas_verified	false	31,011	31,011
0d63	0d63	English	cas_verified	false	31,011	31,011
0dh2	0dh2	English	cas_verified	false	31,011	31,011
0bxg	0bxg	English	cas_verified	false	31,011	31,011
01yn	01yn	English	cas_verified	false	31,011	31,011
0gdc	0gdc	unknown	cas_verified	false	31,011	31,011
0c9c	0c9c	English	cas_verified	false	31,011	31,011
008p	008p	English	cas_verified	false	31,011	31,011
07w4	07w4	English	cas_verified	false	31,011	31,011
03y7	03y7	English	cas_verified	false	31,011	31,011
0ggs	0ggs	Chinese; English	cas_verified	false	31,011	31,011
0d9l	0d9l	English	cas_verified	false	31,011	31,011
058q	058q	English	cas_verified	false	31,011	31,011
05av	05av	English	cas_verified	false	31,011	31,011
0042	0042	Español	cas_verified	false	31,011	31,011
079t	079t	English	cas_verified	false	31,011	31,011
0ahg	0ahg	English	cas_verified	false	31,011	31,011
09aa	09aa	unknown	cas_verified	false	31,011	31,011
01vb	01vb	English	cas_verified	false	31,011	31,011
0cns	0cns	English; Russian	cas_verified	false	31,011	31,011
0c7o	0c7o	English	cas_verified	false	31,011	31,011
06qa	06qa	English	cas_verified	false	31,011	31,011
00jh	00jh	unknown	cas_verified	false	31,011	31,011
038k	038k	English	cas_verified	false	31,011	31,011
04cd	04cd	English	cas_verified	false	31,011	31,011
0gbk	0gbk	unknown	cas_verified	false	31,011	31,011
0cds	0cds	unknown	cas_verified	false	31,011	31,011
0cc5	0cc5	unknown	cas_verified	false	31,011	31,011
03cn	03cn	English	cas_verified	false	31,011	31,011
01gt	01gt	unknown	cas_verified	false	31,011	31,011
01gg	01gg	unknown	cas_verified	false	31,011	31,011
01ku	01ku	English	cas_verified	false	31,011	31,011
039y	039y	English	cas_verified	false	31,011	31,011
0cc6	0cc6	unknown	cas_verified	false	31,011	31,011
0007	0007	español	cas_verified	false	31,011	31,011
08rn	08rn	English	cas_verified	false	31,011	31,011
0bi3	0bi3	unknown	cas_verified	false	31,011	31,011
07nw	07nw	unknown	cas_verified	false	31,011	31,011
0gz6	0gz6	English	cas_verified	false	31,011	31,011
0lcy	0lcy	English	cas_verified	false	31,011	31,011
0ev7	0ev7	unknown	cas_verified	false	31,011	31,011
0doj	0doj	English	cas_verified	false	31,011	31,011
0agi	0agi	English	cas_verified	false	31,011	31,011
0g9j	0g9j	unknown	cas_verified	false	31,011	31,011
04cl	04cl	English	cas_verified	false	31,011	31,011
07v2	07v2	English	cas_verified	false	31,011	31,011
03r8	03r8	English	cas_verified	false	31,011	31,011
09ax	09ax	unknown	cas_verified	false	31,011	31,011
06a8	06a8	unknown	cas_verified	false	31,011	31,011
09vb	09vb	English	cas_verified	false	31,011	31,011
06te	06te	English	cas_verified	false	31,011	31,011
04e0	04e0	unknown	cas_verified	false	31,011	31,011
09m8	09m8	English	cas_verified	false	31,011	31,011
015b	015b	unknown	cas_verified	false	31,011	31,011
06r2	06r2	English	cas_verified	false	31,011	31,011
04rg	04rg	English	cas_verified	false	31,011	31,011
0b6k	0b6k	unknown	cas_verified	false	31,011	31,011
01tj	01tj	English	cas_verified	false	31,011	31,011
0cos	0cos	English; Russian	cas_verified	false	31,011	31,011
06sx	06sx	unknown	cas_verified	false	31,011	31,011
0cp5	0cp5	English; Russian	cas_verified	false	31,011	31,011
0glb	0glb	unknown	cas_verified	false	31,011	31,011
0bwb	0bwb	unknown	cas_verified	false	31,011	31,011
04c1	04c1	English	cas_verified	false	31,011	31,011
03t0	03t0	English	cas_verified	false	31,011	31,011
08y8	08y8	unknown	cas_verified	false	31,011	31,011
0aj0	0aj0	English	cas_verified	false	31,011	31,011
04l0	04l0	unknown	cas_verified	false	31,011	31,011
04t7	04t7	English	cas_verified	false	31,011	31,011
0hh7	0hh7	English	cas_verified	false	31,011	31,011
0e53	0e53	unknown	cas_verified	false	31,011	31,011

End of preview.

RecipeMatching evaluation artifacts

Per-query rank dumps and embedding caches behind the paper Recipe-Matching, Not Equivalence (Ali Habibullah, Mohammad Alshiekh, Yazan Alshoibi, Salman Khan and Naeemullah Khan, 2026). Code, data and results: https://github.com/KAUST-Academy/recipe-matching-not-equivalence.

These files are too large for the GitHub repository. Every number in the paper is already stored there as a result JSON; the files here are what the analysis scripts need in order to recompute the confidence intervals and the style analyses without re-running any model.

Path	Files	Size	What it is
`results/ranks/*.ranks.jsonl`	891	1.8 GB	One JSON line per query: its id and the rank of its gold document (fields vary by evaluation, e.g. `gold_rank`, `gold_sim`, `top10_ids`). Written by `scripts/eval_retrieve.py --dump-ranks`, `scripts/eval_crosslingual.py`, `scripts/eval_bm25.py` and `scripts/colbert_eval.py`. The repository's `results/ranks/*.summary.json` files are their aggregates.
`emb_cache/*.npz`	16	4.4 GB	MathNet-Retrieve document (117,088) and query (15,000) embeddings of eight fine-tuned checkpoints, arrays `ids` and `emb`: `ctrl-cas-6145`, `ctrl-llm-6145`, `dose-paraphrase-6145`, `dose-style-6145`, `dose-unrelated-6145`, `fact-d4casnegs-6145-s42`, `fact-d4llmnegs-6145-s42` and `fact-d4llmnegsfull-6145-s42` (training records in `models/<name>/run_config.json` of the repository).
`upstream_versions/`	2	small	The Hugging Face revision of every base model and dataset the experiments downloaded, recorded on 2026-09-28, and the commit of the MIRB code.

Use

Clone the GitHub repository, then place the files where the scripts expect them:

hf download KAUSTAcademy/RecipeMatching_eval-artifacts --repo-type dataset --local-dir artifacts
cp artifacts/results/ranks/*.ranks.jsonl results/ranks/
mkdir -p .emb_cache && cp artifacts/emb_cache/*.npz .emb_cache/

The bootstrap, difference-in-differences and slice scripts (for example scripts/bootstrap_stats.py, scripts/dose_bootstrap.py, scripts/reference_control_did.py, scripts/xling_slice_analysis.py and scripts/samelang_verdict.py) read the rank dumps; scripts/style_probe.py, scripts/style_within_class.py and scripts/analyze_hits.py read the embeddings.

Citation

@misc{habibullah2026recipematchingequivalence,
      title={Recipe-Matching, Not Equivalence}, 
      author={Ali Habibullah and Mohammad Alshiekh and Yazan Alshoibi and Salman Khan and Naeemullah Khan},
      year={2026},
      eprint={2609.31927},
      archivePrefix={arXiv},
      primaryClass={cs.IR},
      url={https://arxiv.org/abs/2609.31927}, 
}

Downloads last month: 9

Total file size:

6.26 GB

Paper for KAUSTAcademy/RecipeMatching_eval-artifacts

Recipe-Matching, Not Equivalence

Paper • 2609.31927 • Published 8 days ago

RecipeMatching evaluation artifacts

Contents

Use

Citation

Paper for KAUSTAcademy/RecipeMatching_eval-artifacts