training · complete-prefixes-weaker-out-of-family-model-during TEC-0104 · proposed

Complete prefixes from a weaker out-of-family model during R

For a fraction of RLVR rollouts, prepend a truncated partial reasoning trajectory generated by a smaller, different-family model (cut at the largest step-to-step entropy drop) and train the target model to complete it.

Proposed. Filed by the drafting stage because a paper introduces or tests it. Nobody has vouched for it. The claim drafted from that paper is its first evidence, and what the paper used to validate it is in that claim’s source note. Say whether it held up for you, below.

For a fraction of RLVR rollouts, prepend a truncated partial reasoning trajectory generated by a smaller, different-family model (cut at the largest step-to-step entropy drop) and train the target model to complete it. Filed by the drafting stage from Boosting LLM Exploration via Weak-Model Guidance in RLVR (2608.27420) with status proposed: the paper introduces or tests it, and the claim drafted from the paper is its first evidence. The paper validated it by: Average pass@1 to pass@128 over AIME 2024/2025, AMC 2023, MATH 500, Minerva, OlympiadBench against vanilla GRPO and the untrained base model; ablations on injection probability, random vs entropy truncation, and same-family prefix generators; LLM-graded prefix quality distribution.. Nobody has vouched for it; add support or contest it on this page.

Addresses: Retaining skills and safety after fine-tuning

Does it work?

nothing measured0 supporting · 0 contesting sources

Efficacy claims — what this technique actually moves, under which conditions, and whether that has been contested.

No efficacy claim filed yet. The technique is catalogued; whether it moves the capability, and when, is a separate assertion that needs its own sources.

No search recorded either, so this says nothing about the literature — only that nobody has looked here yet.

Code

No repository linked yet. Contribute one.

Sources

Either way, bring evidence: a link, or what you saw.

Whether a technique works is a claim. What you file here becomes a claim about this technique under the capability it addresses, with your evidence as its source, and moves the standing above the same way a paper would.