Skip to content
PostTrainLLM docs
Esc
navigateopen⌘Jpreview
On this page

SQL Technique Lineage — Closed

This is the SQL-specific ledger of methods, recipes, status, and next smallest tests. It exists to prevent the roadmap from looking complete just because broad method names are present.

Current Evidence

Recipe / Attempt Status Evidence Decision
Toy SQL SFT, rank 4 Worked as a factory proof execution 0.167 -> 0.833 on 6 rows retry-data because train/eval overlap
Expanded synthetic SQL SFT, rank 4 Worked execution 0.160 -> 0.860, exact 0.140 -> 0.840 on 50 rows current synthetic specialist
Public b-mc2 SFT v1 Failed exact 0.344 vs T5-small 0.484 reject
Public b-mc2 SFT v2 Failed badly exact 0.031 reject
Public b-mc2 SFT v3 Improved but failed gate exact 0.422 reject
Public b-mc2 SFT v4, join/group weighted Worked on public exact exact 0.531 vs T5-small 0.484 not shippable; synthetic regression
Public v4 synthetic regression Failed synthetic execution 0.240 vs incumbent 0.860 route or compose
Public+synthetic blended SFT Failed/interfered public 0.297, synthetic execution 0.560 reject
Static multi-LoRA composition Failed to pass both gates best public pass still had synthetic failure reject
Routed public + synthetic adapters Best current candidate public exact 0.531, synthetic execution 0.860 report-ready, not package-ready
Hygiene SimPO/DPO adapter Failed hard execution 0.860 -> 0.080, degenerate output retry-training

Closed Historical Proposals

Technique Method Historical proposal Final disposition Fresh-work rule
Candidate-selection curriculum supervised selection / RLVR bridge Build rows with 4-6 candidate SQL strings and learn to pick the executable answer rejected: no frozen candidate dataset or eval existed before project closure Reopen only as a new experiment after learning with leakage checks and a frozen selection gate
Reference-anchored hygiene DPO DPO Retry the failed hygiene goal with a reference anchor, lower LR, fewer steps, and composed eval against the SFT adapter Ref-free SimPO over-optimized and collapsed; reference anchoring should preserve generation quality Same 108 pairs, frozen 50-row eval, require no execution regression and clean-SQL lift
One-step offline rollout update OAPL/ReST-style batch loop Generate N rollouts per prompt, score offline, and train one adapter update rejected: its candidate-selection reward prerequisite was never validated A new experiment must validate the reward before any update
Policy lag / stale reference RL regularization Keep the rollout policy fixed or compare against a stale reference rejected: no valid offline rollout experiment existed to vary Treat only as a variable inside a future fresh RL experiment
Controlled LoRA rank sweep LoRA/DoRA Sweep rank {1,2,4,8} with every other variable fixed rejected: no active frozen SQL target existed, so the sweep would not resolve the historical confound Start as a new experiment with fixed data, seed, steps, LR, and eval
LoRA geometry decision check diagnostics Compare successful, failed, and retry adapters by effective-update norm/stable rank/module concentration Explains whether a failure learned too little, too diffusely, or in the wrong layers Run scripts/lora_geometry.py on every meaningful adapter and attach lora-geometry.json
Slice-gated reporting eval discipline Require overall, join, single-table, filter, aggregate, format, and clean-output slices Overall hides the known join weakness and hygiene failure Generate slice-metrics.json for every SQL report
Trace review as data source failure analysis Classify failures into hallucinated schema, missing join, wrong filter, prose/fence wrapping, no-select collapse Converts failed attempts into targeted data or preferences Generate trace_review.md for every SQL report

Closure

There is no SQL priority queue after this pass. The measured lineage, failed recipes, routed result, and rejected proposals are retained as a learning lab. Future SQL work begins from a newly frozen question rather than continuing this historical sequence by inertia.

Enforcement

No SQL candidate should be reported as improved unless the run folder contains:

  • eval-baseline.json
  • eval-candidate.json
  • slice-metrics.json
  • trace_review.md
  • report.md
  • decision.json

No SQL candidate should be packaged unless the decision is ship and the public execution gate is present or explicitly waived in the report.

Was this page helpful?