Van Nyaya datasets
Collection
CuratorKIT grounded QA (English + Hindi), refusal and DPO data from Indian forest and environment law, plus eval outputs. • 3 items • Updated
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
base: struct<grounded_en_f1: double, grounded_en_rl: double, grounded_en_falseref: double, grounded_en_wor (... 155 chars omitted)
child 0, grounded_en_f1: double
child 1, grounded_en_rl: double
child 2, grounded_en_falseref: double
child 3, grounded_en_words: double
child 4, grounded_hi_f1: double
child 5, grounded_hi_rl: double
child 6, grounded_hi_falseref: double
child 7, grounded_hi_words: double
child 8, unans: double
child 9, n: int64
child 10, empty: int64
base_ours: struct<grounded_en_f1: double, grounded_en_rl: double, grounded_en_falseref: double, grounded_en_wor (... 155 chars omitted)
child 0, grounded_en_f1: double
child 1, grounded_en_rl: double
child 2, grounded_en_falseref: double
child 3, grounded_en_words: double
child 4, grounded_hi_f1: double
child 5, grounded_hi_rl: double
child 6, grounded_hi_falseref: double
child 7, grounded_hi_words: double
child 8, unans: double
child 9, n: int64
child 10, empty: int64
method: string
auditkit_version: string
models: struct<base.jsonl: struct<summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_ (... 631 chars omitted)
child 0, base.jsonl: struct<summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>, (... 249 chars omitted)
child 0, summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>, hi_grounded: str (... 141 chars omitted)
child 0, en_grounded: struct<n: i
...
curacy: double>>
child 0, item: struct<idx: int64, f1: double, expected_action_accuracy: double>
child 0, idx: int64
child 1, f1: double
child 2, expected_action_accuracy: double
child 1, sft.jsonl: struct<summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>, (... 249 chars omitted)
child 0, summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>, hi_grounded: str (... 141 chars omitted)
child 0, en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>
child 0, n: int64
child 1, f1: double
child 2, expected_action_accuracy: double
child 1, hi_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>
child 0, n: int64
child 1, f1: double
child 2, expected_action_accuracy: double
child 2, en_unanswerable: struct<n: int64, f1: double, expected_action_accuracy: double>
child 0, n: int64
child 1, f1: double
child 2, expected_action_accuracy: double
child 1, per_example: list<item: struct<idx: int64, f1: double, expected_action_accuracy: double>>
child 0, item: struct<idx: int64, f1: double, expected_action_accuracy: double>
child 0, idx: int64
child 1, f1: double
child 2, expected_action_accuracy: double
to
{'auditkit_version': Value('string'), 'method': Value('string'), 'models': {'base.jsonl': {'summary': {'en_grounded': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}, 'hi_grounded': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}, 'en_unanswerable': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}}, 'per_example': List({'idx': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')})}, 'sft.jsonl': {'summary': {'en_grounded': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}, 'hi_grounded': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}, 'en_unanswerable': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}}, 'per_example': List({'idx': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')})}}}
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 544, 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 3052, in __iter__
for key, example in ex_iterable:
^^^^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2559, in __iter__
for key, pa_table in self._iter_arrow():
~~~~~~~~~~~~~~~~^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 2584, 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 558, in _iter_arrow
for key, pa_table in iterator:
^^^^^^^^
File "/usr/local/lib/python3.14/site-packages/datasets/iterable_dataset.py", line 440, 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 361, 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 2381, in table_cast
return cast_table_to_schema(table, schema)
File "/usr/local/lib/python3.14/site-packages/datasets/table.py", line 2309, in cast_table_to_schema
raise CastError(
...<3 lines>...
)
datasets.table.CastError: Couldn't cast
base: struct<grounded_en_f1: double, grounded_en_rl: double, grounded_en_falseref: double, grounded_en_wor (... 155 chars omitted)
child 0, grounded_en_f1: double
child 1, grounded_en_rl: double
child 2, grounded_en_falseref: double
child 3, grounded_en_words: double
child 4, grounded_hi_f1: double
child 5, grounded_hi_rl: double
child 6, grounded_hi_falseref: double
child 7, grounded_hi_words: double
child 8, unans: double
child 9, n: int64
child 10, empty: int64
base_ours: struct<grounded_en_f1: double, grounded_en_rl: double, grounded_en_falseref: double, grounded_en_wor (... 155 chars omitted)
child 0, grounded_en_f1: double
child 1, grounded_en_rl: double
child 2, grounded_en_falseref: double
child 3, grounded_en_words: double
child 4, grounded_hi_f1: double
child 5, grounded_hi_rl: double
child 6, grounded_hi_falseref: double
child 7, grounded_hi_words: double
child 8, unans: double
child 9, n: int64
child 10, empty: int64
method: string
auditkit_version: string
models: struct<base.jsonl: struct<summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_ (... 631 chars omitted)
child 0, base.jsonl: struct<summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>, (... 249 chars omitted)
child 0, summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>, hi_grounded: str (... 141 chars omitted)
child 0, en_grounded: struct<n: i
...
curacy: double>>
child 0, item: struct<idx: int64, f1: double, expected_action_accuracy: double>
child 0, idx: int64
child 1, f1: double
child 2, expected_action_accuracy: double
child 1, sft.jsonl: struct<summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>, (... 249 chars omitted)
child 0, summary: struct<en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>, hi_grounded: str (... 141 chars omitted)
child 0, en_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>
child 0, n: int64
child 1, f1: double
child 2, expected_action_accuracy: double
child 1, hi_grounded: struct<n: int64, f1: double, expected_action_accuracy: double>
child 0, n: int64
child 1, f1: double
child 2, expected_action_accuracy: double
child 2, en_unanswerable: struct<n: int64, f1: double, expected_action_accuracy: double>
child 0, n: int64
child 1, f1: double
child 2, expected_action_accuracy: double
child 1, per_example: list<item: struct<idx: int64, f1: double, expected_action_accuracy: double>>
child 0, item: struct<idx: int64, f1: double, expected_action_accuracy: double>
child 0, idx: int64
child 1, f1: double
child 2, expected_action_accuracy: double
to
{'auditkit_version': Value('string'), 'method': Value('string'), 'models': {'base.jsonl': {'summary': {'en_grounded': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}, 'hi_grounded': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}, 'en_unanswerable': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}}, 'per_example': List({'idx': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')})}, 'sft.jsonl': {'summary': {'en_grounded': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}, 'hi_grounded': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}, 'en_unanswerable': {'n': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')}}, 'per_example': List({'idx': Value('int64'), 'f1': Value('float64'), 'expected_action_accuracy': Value('float64')})}}}
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.
Van Nyaya eval outputs
Generations and scores of Tiny Aya Global (baseline) and Van Nyaya checkpoints on the 164 held-out prompts (English grounded QA, Hindi grounded QA, unanswerable refusal). Metrics: token F1, ROUGE-L, refusal rate on unanswerable prompts, false-refusal rate on answerable prompts.
| model | EN F1 | EN ROUGE-L | HI F1 | HI ROUGE-L | refusal on unanswerable | false refusal |
|---|---|---|---|---|---|---|
| Tiny Aya Global (base) | 0.450 | 0.303 | 0.640 | 0.366 | 20.0% | 1.1% |
| Van Nyaya DPO-on-base (step 20) | 0.436 | 0.292 | 0.645 | 0.371 | 37.5% | 2.1% |
More checkpoints (SFT, SFT+DPO) are added as training finishes.
- Downloads last month
- -