---
license: other
license_name: weft-research
pretty_name: Weft Script-Lineage (synthetic + negative-result evidence)
tags:
- data-lineage
- etl
- synthetic
- negative-result
language:
- en
---

# weft-script-lineage — synthetic training data + negative-result evidence

Companion data for the model [wallfacers/weft-lineage-extractor-1.5b](https://huggingface.co/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.