Hugging Face's logo

Datasets:
Jacqkues
/
sev-generic-bench

Tags:
evaluation
vision
calibration
contamination
License:
Dataset card Files
xet
 Community
Dataset Viewer
Duplicate
The dataset viewer is not available for this split.
Cannot load the dataset split (in streaming mode) to extract the first rows.
Error code:   StreamingRowsError
Exception:    CastError
Message:      Couldn't cast
run: string
id: string
bench: string
category: string
label: int64
z: list<item: double>
  child 0, item: double
flagged: struct<mmstar: list<item: struct<id: string, source: string, bits: int64>>, mmbench: list<item: stru (... 257 chars omitted)
  child 0, mmstar: list<item: struct<id: string, source: string, bits: int64>>
      child 0, item: struct<id: string, source: string, bits: int64>
          child 0, id: string
          child 1, source: string
          child 2, bits: int64
  child 1, mmbench: list<item: struct<id: string, source: string, bits: int64>>
      child 0, item: struct<id: string, source: string, bits: int64>
          child 0, id: string
          child 1, source: string
          child 2, bits: int64
  child 2, mme: list<item: struct<id: string, source: string, bits: int64>>
      child 0, item: struct<id: string, source: string, bits: int64>
          child 0, id: string
          child 1, source: string
          child 2, bits: int64
  child 3, realworldqa: list<item: struct<id: string, source: string, bits: int64>>
      child 0, item: struct<id: string, source: string, bits: int64>
          child 0, id: string
          child 1, source: string
          child 2, bits: int64
  child 4, hallusion: list<item: struct<id: string, source: string, bits: int64>>
      child 0, item: struct<id: string, source: string, bits: int64>
          child 0, id: string
          child 1, source: string
          child 2, bits: int64
max_bits: int64
benches: struct<
...
rosat: int64
  child 2, mme: struct<n: int64, flagged: int64, share: double, by_source: struct<vqav2: int64, chartnet: int64, doc (... 87 chars omitted)
      child 0, n: int64
      child 1, flagged: int64
      child 2, share: double
      child 3, by_source: struct<vqav2: int64, chartnet: int64, documents: int64, ava: int64, eurosat: int64, scienceqa: int64 (... 27 chars omitted)
          child 0, vqav2: int64
          child 1, chartnet: int64
          child 2, documents: int64
          child 3, ava: int64
          child 4, eurosat: int64
          child 5, scienceqa: int64
          child 6, docindex/docindex: int64
  child 3, realworldqa: struct<n: int64, flagged: int64, share: double, by_source: struct<vqav2: int64, aokvqa: int64, scien (... 39 chars omitted)
      child 0, n: int64
      child 1, flagged: int64
      child 2, share: double
      child 3, by_source: struct<vqav2: int64, aokvqa: int64, scienceqa: int64, docindex/docindex: int64>
          child 0, vqav2: int64
          child 1, aokvqa: int64
          child 2, scienceqa: int64
          child 3, docindex/docindex: int64
  child 4, hallusion: struct<n: int64, flagged: int64, share: double, by_source: struct<scienceqa: int64, docindex/docinde (... 10 chars omitted)
      child 0, n: int64
      child 1, flagged: int64
      child 2, share: double
      child 3, by_source: struct<scienceqa: int64, docindex/docindex: int64>
          child 0, scienceqa: int64
          child 1, docindex/docindex: int64
to
{'max_bits': Value('int64'), 'benches': {'mmstar': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'aokvqa': Value('int64'), 'vqav2': Value('int64'), 'documents': Value('int64'), 'scienceqa': Value('int64'), 'docindex/docindex': Value('int64'), 'eurosat': Value('int64')}}, 'mmbench': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'vqav2': Value('int64'), 'scienceqa': Value('int64'), 'aokvqa': Value('int64'), 'documents': Value('int64'), 'chartnet': Value('int64'), 'docindex/docindex': Value('int64'), 'eurosat': Value('int64')}}, 'mme': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'vqav2': Value('int64'), 'chartnet': Value('int64'), 'documents': Value('int64'), 'ava': Value('int64'), 'eurosat': Value('int64'), 'scienceqa': Value('int64'), 'docindex/docindex': Value('int64')}}, 'realworldqa': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'vqav2': Value('int64'), 'aokvqa': Value('int64'), 'scienceqa': Value('int64'), 'docindex/docindex': Value('int64')}}, 'hallusion': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'scienceqa': Value('int64'), 'docindex/docindex': Value('int64')}}}, 'flagged': {'mmstar': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')}), 'mmbench': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')}), 'mme': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')}), 'realworldqa': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')}), 'hallusion': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')})}}
because column names don't match
Traceback:    Traceback (most recent call last):
                File "/src/services/worker/src/worker/utils.py", line 147, in get_rows_or_raise
                  return get_rows(
                      dataset=dataset,
                  ...<4 lines>...
                      column_names=column_names,
                  )
                File "/src/libs/libcommon/src/libcommon/utils.py", line 272, in decorator
                  return func(*args, **kwargs)
                File "/src/services/worker/src/worker/utils.py", line 127, in get_rows
                  rows_plus_one = list(itertools.islice(safe_iter(ds, dataset=dataset), rows_max_number + 1))
                File "/src/services/worker/src/worker/utils.py", line 483, in safe_iter
                  yield from ds.decode(False) if ds.features else ds
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2840, in __iter__
                  for key, example in ex_iterable:
                                      ^^^^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2373, in __iter__
                  for key, pa_table in self._iter_arrow():
                                       ~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2398, in _iter_arrow
                  for key, pa_table in self.ex_iterable._iter_arrow():
                                       ~~~~~~~~~~~~~~~~~~~~~~~~~~~~^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 536, in _iter_arrow
                  for key, pa_table in iterator:
                                       ^^^^^^^^
                File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 419, in _iter_arrow
                  for key, pa_table in self.generate_tables_fn(**gen_kwags):
                                       ~~~~~~~~~~~~~~~~~~~~~~~^^^^^^^^^^^^^
                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
              run: string
              id: string
              bench: string
              category: string
              label: int64
              z: list<item: double>
                child 0, item: double
              flagged: struct<mmstar: list<item: struct<id: string, source: string, bits: int64>>, mmbench: list<item: stru (... 257 chars omitted)
                child 0, mmstar: list<item: struct<id: string, source: string, bits: int64>>
                    child 0, item: struct<id: string, source: string, bits: int64>
                        child 0, id: string
                        child 1, source: string
                        child 2, bits: int64
                child 1, mmbench: list<item: struct<id: string, source: string, bits: int64>>
                    child 0, item: struct<id: string, source: string, bits: int64>
                        child 0, id: string
                        child 1, source: string
                        child 2, bits: int64
                child 2, mme: list<item: struct<id: string, source: string, bits: int64>>
                    child 0, item: struct<id: string, source: string, bits: int64>
                        child 0, id: string
                        child 1, source: string
                        child 2, bits: int64
                child 3, realworldqa: list<item: struct<id: string, source: string, bits: int64>>
                    child 0, item: struct<id: string, source: string, bits: int64>
                        child 0, id: string
                        child 1, source: string
                        child 2, bits: int64
                child 4, hallusion: list<item: struct<id: string, source: string, bits: int64>>
                    child 0, item: struct<id: string, source: string, bits: int64>
                        child 0, id: string
                        child 1, source: string
                        child 2, bits: int64
              max_bits: int64
              benches: struct<
              ...
              rosat: int64
                child 2, mme: struct<n: int64, flagged: int64, share: double, by_source: struct<vqav2: int64, chartnet: int64, doc (... 87 chars omitted)
                    child 0, n: int64
                    child 1, flagged: int64
                    child 2, share: double
                    child 3, by_source: struct<vqav2: int64, chartnet: int64, documents: int64, ava: int64, eurosat: int64, scienceqa: int64 (... 27 chars omitted)
                        child 0, vqav2: int64
                        child 1, chartnet: int64
                        child 2, documents: int64
                        child 3, ava: int64
                        child 4, eurosat: int64
                        child 5, scienceqa: int64
                        child 6, docindex/docindex: int64
                child 3, realworldqa: struct<n: int64, flagged: int64, share: double, by_source: struct<vqav2: int64, aokvqa: int64, scien (... 39 chars omitted)
                    child 0, n: int64
                    child 1, flagged: int64
                    child 2, share: double
                    child 3, by_source: struct<vqav2: int64, aokvqa: int64, scienceqa: int64, docindex/docindex: int64>
                        child 0, vqav2: int64
                        child 1, aokvqa: int64
                        child 2, scienceqa: int64
                        child 3, docindex/docindex: int64
                child 4, hallusion: struct<n: int64, flagged: int64, share: double, by_source: struct<scienceqa: int64, docindex/docinde (... 10 chars omitted)
                    child 0, n: int64
                    child 1, flagged: int64
                    child 2, share: double
                    child 3, by_source: struct<scienceqa: int64, docindex/docindex: int64>
                        child 0, scienceqa: int64
                        child 1, docindex/docindex: int64
              to
              {'max_bits': Value('int64'), 'benches': {'mmstar': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'aokvqa': Value('int64'), 'vqav2': Value('int64'), 'documents': Value('int64'), 'scienceqa': Value('int64'), 'docindex/docindex': Value('int64'), 'eurosat': Value('int64')}}, 'mmbench': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'vqav2': Value('int64'), 'scienceqa': Value('int64'), 'aokvqa': Value('int64'), 'documents': Value('int64'), 'chartnet': Value('int64'), 'docindex/docindex': Value('int64'), 'eurosat': Value('int64')}}, 'mme': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'vqav2': Value('int64'), 'chartnet': Value('int64'), 'documents': Value('int64'), 'ava': Value('int64'), 'eurosat': Value('int64'), 'scienceqa': Value('int64'), 'docindex/docindex': Value('int64')}}, 'realworldqa': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'vqav2': Value('int64'), 'aokvqa': Value('int64'), 'scienceqa': Value('int64'), 'docindex/docindex': Value('int64')}}, 'hallusion': {'n': Value('int64'), 'flagged': Value('int64'), 'share': Value('float64'), 'by_source': {'scienceqa': Value('int64'), 'docindex/docindex': Value('int64')}}}, 'flagged': {'mmstar': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')}), 'mmbench': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')}), 'mme': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')}), 'realworldqa': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')}), 'hallusion': List({'id': Value('string'), 'source': Value('string'), 'bits': Value('int64')})}}
              because column names don't match

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.

Sev generic benchmark

A generic evaluation of typed, calibrated vision decisions. It is built from five public VLM benchmarks: MMStar, MMBench-en dev (1,000 sampled items), MME, RealWorldQA and HallusionBench (image split). Every item is cast as a TypeSafe-style question: a choice over the lettered options, or a noul (yes/no).

This repository ships no images. The source benchmarks carry their own licences, so it contains the recipe to rebuild the items instead:

  • generic.py: the deterministic builder and evaluator;
  • contamination.py: the near-duplicate check against the training data;
  • contamination.json: the 305 items flagged in that check;
  • predictions.jsonl: per-item logits of each model;
  • report.json: results on all items and on the clean items, with bootstrap intervals.

Protocol

  • Filtering:
    • ScienceQA-sourced items are dropped, because Sev trained on ScienceQA.
    • Items whose answer is neither a letter nor yes/no are dropped.
  • Contamination control: every benchmark image is hashed with a 256-bit average hash and compared to every training image of kev-vision-decisions-full and doc-index. 305 of the 6,001 items are within 6 bits of a training image and are reported separately.
  • Resolution: all models see images capped at 448 × 448 pixels.
  • Baseline: Qwen3.5-0.8B in zero-shot, scored by its next-token probability over the candidate letters (or Yes/No).

Results

Accuracy on the 5,696 clean items (ECE in parentheses):

benchmark n Qwen3.5-0.8B Sev-0.8B Sev-0.8B-docindex
MMStar 1,036 0.460 (0.18) 0.526 (0.08) 0.516 (0.20)
MMBench dev 927 0.774 (0.02) 0.825 (0.06) 0.812 (0.04)
MME 2,206 0.711 (0.12) 0.771 (0.01) 0.754 (0.10)
RealWorldQA 586 0.570 (0.15) 0.631 (0.05) 0.631 (0.09)
HallusionBench 941 0.644 (0.15) 0.629 (0.10) 0.646 (0.17)
all 5,696 0.650 (0.12) 0.698 (0.02) 0.690 (0.12)

Sev-0.8B's lead over the zero-shot baseline is +4.7 points (paired bootstrap 95% CI [+3.5, +6.0]). HallusionBench is the one benchmark where the difference is not significant.

Rebuild 

uv run modal run modal_app.py::generic_build                 # -> /vol/data/generic.parquet
KEV_GPU=L4 uv run modal run modal_app.py::generic_eval       # -> /vol/runs/eval/generic.json (+ .preds.jsonl)
Downloads last month
23
Total file size:
2.98 MB

Models trained or fine-tuned on Jacqkues/sev-generic-bench