File size: 6,480 Bytes
8125566
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b018641
8125566
b018641
 
 
8125566
b018641
8125566
 
 
b018641
8125566
b018641
 
 
 
 
 
 
8125566
b018641
8125566
 
 
 
 
 
 
b018641
8125566
b018641
 
 
 
 
8125566
b018641
8125566
b018641
8125566
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
b018641
8125566
b018641
8125566
b018641
 
 
 
 
8125566
b018641
8125566
b018641
8125566
 
 
 
 
 
 
 
 
 
 
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
---
license: cc-by-sa-4.0
language:
- ar
- en
task_categories:
- visual-question-answering
- document-question-answering
- image-to-text
pretty_name: RasmBench
size_categories:
- 1K<n<10K
tags:
- arabic
- charts
- tables
- document-understanding
- counterfactual
configs:
- config_name: documents
  default: true
  data_files:
  - split: test
    path: data/documents/test-*.parquet
- config_name: paired_rerenders
  data_files:
  - split: factual_values
    path: data/paired_rerenders/factual_values-*.parquet
  - split: counterfactual_values
    path: data/paired_rerenders/counterfactual_values-*.parquet
- config_name: figure_as_text
  data_files:
  - split: image_table
    path: data/figure_as_text/image_table-*.parquet
  - split: image_table_control
    path: data/figure_as_text/image_table_control-*.parquet
  - split: table_only
    path: data/figure_as_text/table_only-*.parquet
  - split: own_table
    path: data/figure_as_text/own_table-*.parquet
- config_name: translation
  data_files:
  - split: ar_doc_ar_q
    path: data/translation/ar_doc_ar_q-*.parquet
  - split: ar_doc_en_q
    path: data/translation/ar_doc_en_q-*.parquet
  - split: en_doc_ar_q
    path: data/translation/en_doc_ar_q-*.parquet
  - split: en_doc_en_q
    path: data/translation/en_doc_en_q-*.parquet
- config_name: chart_to_table
  data_files:
  - split: selection
    path: data/chart_to_table/selection-*.parquet
  - split: pairs
    path: data/chart_to_table/pairs-*.parquet
  - split: tiers
    path: data/chart_to_table/tiers-*.parquet
- config_name: model_outputs
  data_files:
  - split: test
    path: data/model_outputs/test.parquet
- config_name: second_judge
  data_files:
  - split: test
    path: data/second_judge/test.parquet
---

# RasmBench

1,002 rendered Arabic documents with charts and tables, each with six questions (one per Bloom level) in multiple-choice and open-ended form: 6,012 items per format. Documents are built from Arabic Wikipedia tables in three value regimes (334 each) and five degradation tiers (A born-digital, B print scan, C complex-layout scan, D camera capture, E historical print).

- **factual**: the real values.
- **counterfactual**: every number perturbed, with totals, percentages and year order restored. Labels stay real.
- **neutral**: placeholder labels and random values.

> **Counterfactual items contain deliberately false statistics about real entities** (`values_altered = true`). Do not use them as factual data or train on them as real data.

## Configurations

| Config | Splits (rows) | Contents |
|---|---|---|
| `documents` (default) | `test` (1,002) | The benchmark: page image, undegraded image, six questions |
| `paired_rerenders` | `factual_values`, `counterfactual_values` (668 each) | Each factual/counterfactual document drawn with both value sets, questions re-keyed |
| `figure_as_text` | `image_table` (100), `image_table_control` (50), `table_only` (100), `own_table` (230) | Figure data as a markdown table: with the image, without it, or as the model's own transcription |
| `translation` | `ar_doc_ar_q`, `ar_doc_en_q`, `en_doc_ar_q`, `en_doc_en_q` (98 each) | Arabic/English document × Arabic/English question |
| `chart_to_table` | `selection` (46), `pairs` (97), `tiers` (245) | Transcribe the figure; the reference is what the chart draws |
| `model_outputs` | `test` (201,593) | Every model answer in every condition, with judge scores |
| `second_judge` | `test` (4,252) | mistral-medium-3 as a second judge on 598 sampled questions |

Diagnostic conditions use 100 selected documents (the 50 easiest and 50 hardest across ten models) and 50 random control documents, listed in `metadata/` and in the `selection` column. Full page annotations are in `annotations/` (re-rendered variants also carry theirs in an `annotation` column).

```python
from datasets import load_dataset
docs = load_dataset("rasmbench/rasmbench_revised", split="test")
outputs = load_dataset("rasmbench/rasmbench_revised", "model_outputs", split="test")
```

## Notes

- **Key on `uid`** (`<regime>/<doc_id>`), not `doc_id`: 21 `doc_id`s occur in more than one regime. In variant configs `uid` starts with the condition name, and `source_uid` points to the released document.
- **Drop `distractor_pool` before prompting**; it leaks the answer space.
- The open-ended reference answer is `options[answer_index]`.
- `tier` is the degradation applied. `template_nominal_tier` is the template's design tier and differs on 168 documents.
- `model_outputs`: `condition` is `original` or the variant name, and `document_uid` joins to the variant's `uid`. Unparsed MCQ answers have a null `predicted_index` and count as wrong. Original open-ended answers were judged by gemini-3.5-flash and later ones by gemini-3.8-flash, so compare scores within one `judge_model`.

## Prompts

MCQ, with options as `A) …` to `D) …`:

```
اقرأ الوثيقة في الصورة وأجب عن السؤال التالي.

{question}

{options}

أجب بحرف الخيار الصحيح فقط (A أو B أو C أو D).
```

Open-ended:

```
اقرأ الوثيقة في الصورة وأجب عن السؤال التالي بإيجاز.

{question}
```

Greedy decoding with at most 128 new tokens for MCQ, 256 for open-ended answers and 1,024 for chart-to-table transcription. The judge rubric and chart-to-table prompts are in the paper's appendix.

## Limitations

- Questions were written by gemini-2.5-flash-lite and open-ended answers judged by Gemini, which may favour Gemini-like models.
- Re-rendered questions were re-keyed automatically; between 1 and 3 of 50 hand-checked were wrong.
- English documents are machine translations. 2 of the 100 selected failed to translate (`metadata/translation_failed.json`), and Qwen3.6-27B, gemma-4-31B, Qwen3-VL-4B and Qwen3.5-9B answered 76 of the 98.
- Tiers were not assigned at random, so tier comparisons are observational.
- The rendering engine is not part of this release.

## Licence

CC BY-SA 4.0, derived from Arabic Wikipedia (CC BY-SA 4.0). Credit Arabic Wikipedia and this dataset, and keep the licence. Questions, translations and judge scores were generated with Gemini models and answers with the listed models, so their providers' terms may also apply.

## Citation

```bibtex
@misc{rasmbench,
  title  = {RasmBench: Reading, Not Reasoning, Limits Vision-Language Models on Arabic Charts and Tables},
  author = {Anonymous},
  year   = {2026},
  note   = {Under review}
}
```