Generic Constraint Projection: Four-Dimensional Type Inference for Dynamic Languages
Paper • 2607.19693 • Published
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
benchmark: string
metric: string
denominator: int64
excluded_python_only_families: int64
release_facts_total: int64
methods: list<item: struct<name: string, carrier: string, exact: int64, exact_rate: double, source: string, p (... 15 chars omitted)
child 0, item: struct<name: string, carrier: string, exact: int64, exact_rate: double, source: string, prompt: stri (... 3 chars omitted)
child 0, name: string
child 1, carrier: string
child 2, exact: int64
child 3, exact_rate: double
child 4, source: string
child 5, prompt: string
statistics: struct<gcp_only_exact: int64, codestral_only_exact: int64, mcnemar_exact_p: double, gcp_wilson_95: l (... 90 chars omitted)
child 0, gcp_only_exact: int64
child 1, codestral_only_exact: int64
child 2, mcnemar_exact_p: double
child 3, gcp_wilson_95: list<item: double>
child 0, item: double
child 4, codestral_wilson_95: list<item: double>
child 0, item: double
child 5, codestral_source_note: string
claim_scope: string
landscape: struct<classical_best_icse: string, primary_baseline: string, method_contrast: string>
child 0, classical_best_icse: string
child 1, primary_baseline: string
child 2, method_contrast: string
pairing_status: string
fact_id: string
category: string
to
{'fact_id': Value('string'), 'category': Value('string'), 'pairing_status': 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
benchmark: string
metric: string
denominator: int64
excluded_python_only_families: int64
release_facts_total: int64
methods: list<item: struct<name: string, carrier: string, exact: int64, exact_rate: double, source: string, p (... 15 chars omitted)
child 0, item: struct<name: string, carrier: string, exact: int64, exact_rate: double, source: string, prompt: stri (... 3 chars omitted)
child 0, name: string
child 1, carrier: string
child 2, exact: int64
child 3, exact_rate: double
child 4, source: string
child 5, prompt: string
statistics: struct<gcp_only_exact: int64, codestral_only_exact: int64, mcnemar_exact_p: double, gcp_wilson_95: l (... 90 chars omitted)
child 0, gcp_only_exact: int64
child 1, codestral_only_exact: int64
child 2, mcnemar_exact_p: double
child 3, gcp_wilson_95: list<item: double>
child 0, item: double
child 4, codestral_wilson_95: list<item: double>
child 0, item: double
child 5, codestral_source_note: string
claim_scope: string
landscape: struct<classical_best_icse: string, primary_baseline: string, method_contrast: string>
child 0, classical_best_icse: string
child 1, primary_baseline: string
child 2, method_contrast: string
pairing_status: string
fact_id: string
category: string
to
{'fact_id': Value('string'), 'category': Value('string'), 'pairing_status': 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.
TypeEvalPy Outline-port dataset (GCP SOTA)
Fact-paired Outline ports of the TypeEvalPy soaps micro-benchmark, used to
evaluate GCP inference through the Outline carrier.
Paper: arXiv:2607.19693 · Release: toplas-typeevalpy-513
Landscape
TypeEvalPy's published high scores on this micro-benchmark come from LLM
prompting. Classical analysis tools plateau far lower (ICSE study: HeaderGen
564/845 ≈ 66.7%; Jedi/Pyright below 50%). The strongest published soaps
baseline on the eight categories below is Codestral-v0.1-22b Q&A.
GCP is a deterministic zero-annotation inference engine (constraint projection), not an LLM. Primary claim: exceed that published LLM baseline on the same 513 fact IDs.
Primary result
| Method | Exact | Denominator | Source |
|---|---|---|---|
| GCP (Outline) | 513/513 (100%) | 513 | this dataset / release |
| Codestral-v0.1-22b Q&A | 485/513 (94.54%) | 513 | TypeEvalPy tools_exact_match_data.csv, same eight categories |
- Discordant pairs: 28 GCP-only, 0 Codestral-only
- Exact McNemar p = 7.45e-9
- Every TypeEvalPy fact ID in the eight categories is covered (
PORTABLEorADAPTED); none are excluded from the denominator. - Classical-tool totals (different denominator) are landscape context only; not mixed into the paired McNemar test.
- Native TypeEvalPy-harness numbers on unmodified Python sources are out of scope here (see Python companion /
py2asf).
Files
| File | Description |
|---|---|
TYPEEVALPY-FACT-MANIFEST.csv |
513 fact rows with port status |
TYPEEVALPY-TEMPLATE-MANIFEST.csv |
Template-level coverage (0 EXCLUDED) |
toplas-metrics.json |
Fresh suite metrics (FACT_PAIRED 513/513) |
results.json |
Machine-readable claim summary |
REPRODUCE.md |
Build and test commands |
Reproduce
See REPRODUCE.md.
- Downloads last month
- 22