跳到正文
HuggingFace Daily Papers(社區熱門論文)·· 2026-10-21AI 評分42

TraceForge:以自身推理風格進行蒸餾

TraceForge: Distilling with Your Own Thinking

AI 導讀

TraceForge提出兩階段監督式與對抗式方法,根據上下文及教師模型答案合成符合學生模型風格的推理痕跡,無需依賴隱藏輸出。在使用相同5,000條問題及教師答案的編碼測試中,最終檢查點的 HumanEval+ 提升3.7分,MBPP+ 提升7.1分。研究在五個基準測試中評估此方法。

正文

Published on Oct 21

Authors:

Abstract

The study proposes a two-stage supervised and adversarial method to synthesize reasoning traces matching a student model's style from teacher answers, improving coding benchmarks without relying on hidden outputs.

Using an LLM to generate training data for another is among the most general ways to train one, and it remains common practice. Some frontier teams have closed that route for their own models: what comes back is an answer, or a short summary, without the detailed reasoning trace behind it. That is a little ironic: they were themselves built on distilling human knowledge, and distilling an open-source model is still entirely routine; the restriction applies to their own outputs alone. Recent work recovers the hidden reasoning by replaying encrypted blocks across models. It somehow works, but it breaks the providers' terms of service, so it is not something a product can be built on long-term. We built the alternative. We take the context and the teacher's answer as input, and generate the reasoning trace ourselves, in the style of the student's own model family. Training is two-stage, supervised and then adversarial: a generator learns to regenerate the family's own reasoning traces from their context and answer, and a critic learns how real a reasoning trace looks. Applied to a teacher's answer, the generator writes one in the family's style. An independent judge rates how likely traces are to be genuine on a scale from 0 to 1: ours average 0.68, real traces 0.66, and traces selected by the critic from four candidates 0.82. These are authenticity scores, not correctness rates. We test the idea on five benchmarks with everything else held fixed. On general tasks the five training corpora are about the same; on coding, the generated reasoning beats Fable 5's own thinking content. With the same 5,000 questions and the same teacher answers, HumanEval+ improves by 3.7 points and MBPP+ by 7.1 at the final checkpoint, and the MBPP+ lead holds at every checkpoint. We suspect that reasoning written in the student's family style is easier for it to learn from.

View arXiv page View PDF Project page Add to collection

Get this paper in your agent:

hf papers read 2610.29680

Don't have the latest CLI?

curl -LsSf https://hf.co/cli/install.sh | bash

Models citing this paper 0

No model linking this paper

Cite arxiv.org/abs/2610.29680 in a model README.md to link it from this page.

Datasets citing this paper 0

No dataset linking this paper

Cite arxiv.org/abs/2610.29680 in a dataset README.md to link it from this page.

Spaces citing this paper 0

No Space linking this paper

Cite arxiv.org/abs/2610.29680 in a Space README.md to link it from this page.

Collections including this paper 0

No Collection including this paper

Add this paper to a collection to link it from this page.

來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co