MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning
Paper • 2601.21468 • Published • 25
Dataset Viewer
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
layout_accuracy_pct: struct<1024: double, 64: double, 16: double, 8: double>
child 0, 1024: double
child 1, 64: double
child 2, 16: double
child 3, 8: double
no_layout_accuracy_pct: struct<1024: double, 64: double, 16: double, 8: double>
child 0, 1024: double
child 1, 64: double
child 2, 16: double
child 3, 8: double
layout_8_equals_layout_64: bool
layout_advantage_at_8_points: double
layout_advantage_at_16_points: double
layout_advantage_at_64_points: double
claim4_probe_decision: string
reason: string
poster_html: string
overall: string
canvas: struct<source: string, width_cm: double, height_cm: double, orientation: string, source_url: null>
child 0, source: string
child 1, width_cm: double
child 2, height_cm: double
child 3, orientation: string
child 4, source_url: null
hard_failures: int64
timestamp: timestamp[s]
skill: string
gates: list<item: struct<name: string, severity: string, status: string, command: list<item: string>, summa (... 283 chars omitted)
child 0, item: struct<name: string, severity: string, status: string, command: list<item: string>, summary: struct< (... 271 chars omitted)
child 0, name: string
child 1, severity: string
child 2, status: string
child 3, command: list<item: string>
child 0, item: string
child 4, summary: struct<exit_code: int64, tail: string, gate: string, status: string, rules: list<item: struct<id: in (... 146 chars omitted)
child 0, exit_code: int64
child 1, tail: string
child 2, gate: string
child 3, status: string
child 4, rules: list<item: struct<id: int64, severity: string, status: string, detail: string>>
child 0, item: struct<id: int64, severity: string, status: string, detail: string>
child 0, id: int64
child 1, severity: string
child 2, status: string
child 3, detail: string
child 5, checks: list<item: struct<id: string, severity: string, status: string, detail: string>>
child 0, item: struct<id: string, severity: string, status: string, detail: string>
child 0, id: string
child 1, severity: string
child 2, status: string
child 3, detail: string
child 5, artifacts: list<item: string>
child 0, item: string
warnings: int64
schema_version: int64
to
{'schema_version': Value('int64'), 'skill': Value('string'), 'timestamp': Value('timestamp[s]'), 'poster_html': Value('string'), 'canvas': {'source': Value('string'), 'width_cm': Value('float64'), 'height_cm': Value('float64'), 'orientation': Value('string'), 'source_url': Value('null')}, 'overall': Value('string'), 'hard_failures': Value('int64'), 'warnings': Value('int64'), 'gates': List({'name': Value('string'), 'severity': Value('string'), 'status': Value('string'), 'command': List(Value('string')), 'summary': {'exit_code': Value('int64'), 'tail': Value('string'), 'gate': Value('string'), 'status': Value('string'), 'rules': List({'id': Value('int64'), 'severity': Value('string'), 'status': Value('string'), 'detail': Value('string')}), 'checks': List({'id': Value('string'), 'severity': Value('string'), 'status': Value('string'), 'detail': Value('string')})}, 'artifacts': List(Value('string'))})}
because column names don't match
Traceback: Traceback (most recent call last):
File "/src/services/worker/src/worker/utils.py", line 149, 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 129, 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 489, 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 2818, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2355, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2380, 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 2369, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2297, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
layout_accuracy_pct: struct<1024: double, 64: double, 16: double, 8: double>
child 0, 1024: double
child 1, 64: double
child 2, 16: double
child 3, 8: double
no_layout_accuracy_pct: struct<1024: double, 64: double, 16: double, 8: double>
child 0, 1024: double
child 1, 64: double
child 2, 16: double
child 3, 8: double
layout_8_equals_layout_64: bool
layout_advantage_at_8_points: double
layout_advantage_at_16_points: double
layout_advantage_at_64_points: double
claim4_probe_decision: string
reason: string
poster_html: string
overall: string
canvas: struct<source: string, width_cm: double, height_cm: double, orientation: string, source_url: null>
child 0, source: string
child 1, width_cm: double
child 2, height_cm: double
child 3, orientation: string
child 4, source_url: null
hard_failures: int64
timestamp: timestamp[s]
skill: string
gates: list<item: struct<name: string, severity: string, status: string, command: list<item: string>, summa (... 283 chars omitted)
child 0, item: struct<name: string, severity: string, status: string, command: list<item: string>, summary: struct< (... 271 chars omitted)
child 0, name: string
child 1, severity: string
child 2, status: string
child 3, command: list<item: string>
child 0, item: string
child 4, summary: struct<exit_code: int64, tail: string, gate: string, status: string, rules: list<item: struct<id: in (... 146 chars omitted)
child 0, exit_code: int64
child 1, tail: string
child 2, gate: string
child 3, status: string
child 4, rules: list<item: struct<id: int64, severity: string, status: string, detail: string>>
child 0, item: struct<id: int64, severity: string, status: string, detail: string>
child 0, id: int64
child 1, severity: string
child 2, status: string
child 3, detail: string
child 5, checks: list<item: struct<id: string, severity: string, status: string, detail: string>>
child 0, item: struct<id: string, severity: string, status: string, detail: string>
child 0, id: string
child 1, severity: string
child 2, status: string
child 3, detail: string
child 5, artifacts: list<item: string>
child 0, item: string
warnings: int64
schema_version: int64
to
{'schema_version': Value('int64'), 'skill': Value('string'), 'timestamp': Value('timestamp[s]'), 'poster_html': Value('string'), 'canvas': {'source': Value('string'), 'width_cm': Value('float64'), 'height_cm': Value('float64'), 'orientation': Value('string'), 'source_url': Value('null')}, 'overall': Value('string'), 'hard_failures': Value('int64'), 'warnings': Value('int64'), 'gates': List({'name': Value('string'), 'severity': Value('string'), 'status': Value('string'), 'command': List(Value('string')), 'summary': {'exit_code': Value('int64'), 'tail': Value('string'), 'gate': Value('string'), 'status': Value('string'), 'rules': List({'id': Value('int64'), 'severity': Value('string'), 'status': Value('string'), 'detail': Value('string')}), 'checks': List({'id': Value('string'), 'severity': Value('string'), 'status': Value('string'), 'detail': Value('string')})}, 'artifacts': List(Value('string'))})}
because column names don't matchNeed help to make the dataset viewer work? Make sure to review how to configure the dataset viewer, and open a discussion for direct support.
MemOCR ICML 2026 reproduction artifacts
Scripts and machine-readable outputs for a scaled reproduction of MemOCR: Layout-Aware Visual Memory for Efficient Long-Horizon Reasoning (arXiv:2601.21468; OpenReview vv4TMkyfWA).
The primary GPU script, hf_gpu_layout_probe.py, evaluates the released meituan/MemOCR-7B checkpoint on a deterministic four-example subset of HotpotQA and 2WikiMultiHopQA with approximately 10K-token contexts. It compares layout-preserving and layout-stripped visual memories at 1024/64/16/8 visual-token budgets. This is explicitly a scaled mechanism probe, not the paper's full 4-dataset, 3-context-length, 3-seed evaluation.
- Downloads last month
- 121