product review

Castform RL fine-tune platform

Composite rewards, trace-driven data loops, reasoning-depth classification, and explicit environment contracts make reinforcement-style fine-tuning more inspectable. The hosted dashboard and pay-per-compute product model were not copied, and the scaffolding is not proof that an RLVR run improves a specialist.

Disposition
partially adopted
Study class
product review
Reviewed
2026-09-23
Evidence
Tracked source record

What mattered

Composite rewards, trace-driven data loops, reasoning-depth classification, and explicit environment contracts make reinforcement-style fine-tuning more inspectable.

What changed locally

PostTrainLLM mapped those ideas into composite reward scaffolding, trace-to-data workflows, and reasoning-depth classification.

Evidence limits

The hosted dashboard and pay-per-compute product model were not copied, and the scaffolding is not proof that an RLVR run improves a specialist.

Retained disposition

Connect rewards to an authorized training loop only after the evaluator and held-out gate are frozen.

The action is retained as study context rather than an active backlog item. Any implementation, download, training run, or benchmark requires a fresh scoped question under the repository's experiment gate.

How to read this dossier

This canonical page separates the retained repository decision from the material that informed it. A source may describe an external claim, historical measurement, local observation, or planned procedure; those evidence classes are not interchangeable. Follow the provenance links before reusing a number or method.

Status and disposition describe what PostTrainLLM retained when this record was indexed. They are not a live product promise, a newly run benchmark, or permission to restart historical work. Unknown or unmeasured fields remain unknown rather than being treated as zero.

To reuse the record, first name the exact claim you need and trace it to the linked source. Then check whether the original environment, model revision, data split, evaluator, hardware, and budget match the proposed use. If they do not, treat the record as a hypothesis or design reference and run the smallest fresh comparison that can falsify it. Preserve negative outcomes and regressions beside any improvement; a local win on one slice does not silently become a general capability claim.

Source provenance

The normalized record comes from docs/studies/registry.json. The links below are the tracked evidence and explanatory sources preserved with the record.