wallfacers/weft-lineage-extractor-1.5b
Text Generation • 2B • Updated • 569
Dataset Viewer
The dataset could not be loaded because the splits use different data file formats, which is not supported. Read more about the splits configuration. Click for more details.
Couldn't infer the same data file format for all splits. Got {NamedSplit('train'): ('json', {}), NamedSplit('test'): (None, {})}
Error code: FileFormatMismatchBetweenSplitsError
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.
weft-script-lineage — synthetic training data + negative-result evidence
Companion data for the model wallfacers/weft-lineage-extractor-1.5b, a research artifact demonstrating that synthetic-only training induces a memorization leak in small models for ETL table-lineage extraction.
Contents
train.jsonl/heldout.jsonl— 10,000 + 600 synthetic ETL scripts with table-lineage labels (Python/Shell).out-jvm/adds the Scala/Java-augmented variant.reports/— the frozen evaluation evidence:eval-report-v2.md— synthetic held-out (near-perfect, the misleading number).eval-real.md/eval-real-jvm.md— four-way comparison on real GitHub scripts.leak-report.md/leak-report-jvm.md— memorization-leak quantification.leak-curve.md— leak vs. model scale (0.5B/1.5B/3B).paper-negative-result-findings.md— the full write-up.ADJUDICATION.md— the human-gold labeling convention (Convention A) + provenance.
Note on real scripts
The real-evaluation gold (realeval/gold/*.jsonl) contains third-party GitHub script
content and is not included here by default (license/redistribution care). It is uploaded
only when the publisher passes --include-real-gold after license review.
- Downloads last month
- 42