CoolFace
Datasetpublic

wallfacers/weft-script-lineage-synth

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… See the full description on the dataset page: https://huggingface.co/datasets/wallfacers/weft-script-lineage-synth.

sourceHugging Faceotherupdated 3mo agoView on Hugging Face
0likes38downloads
13 commits on main
4bd371a3mo ago

Upload reports/paper-negative-result-findings.md with huggingface_hub

wallfacers
9c291f03mo ago

Upload reports/ADJUDICATION.md with huggingface_hub

wallfacers
a9bc4253mo ago

Upload reports/leak-curve.md with huggingface_hub

wallfacers
e4292353mo ago

Upload reports/leak-report-jvm.md with huggingface_hub

wallfacers
ae47daf3mo ago

Upload reports/leak-report.md with huggingface_hub

wallfacers
afea41e3mo ago

Upload reports/eval-real-jvm.md with huggingface_hub

wallfacers
2fb90c93mo ago

Upload reports/eval-real.md with huggingface_hub

wallfacers
882b73e3mo ago

Upload reports/eval-report-v2.md with huggingface_hub

wallfacers
5e410433mo ago

041-R v2: JVM-augmented synthetic

wallfacers
2e80e703mo ago

Upload README.md with huggingface_hub

wallfacers
58f95a43mo ago

Upload eval-report.md with huggingface_hub

wallfacers
2f1bbbc3mo ago

041 v2: train+heldout synth data

wallfacers
5bc775a3mo ago

initial commit

wallfacers