arXiv AI October 16, 2026 By Feiyang Chen, Runze Ma, Zichen Wang, Jiong Lou, Zihan Wang, Shuangjia Zheng, Jie Li

Learning from Unverified Peers via Shared Residual Distillation

The paper introduces Shared Residual Distillation (SRD), an on‑policy self‑distillation technique that learns from a language model’s own unverified peer solutions without requiring external supervision. SRD compares how peers alter the prediction distribution relative to a control, keeping only adjustments that move in the same direction and are conservative in magnitude. Experiments show that SRD significantly boosts performance on the AIME24–26 benchmark for Qwen3 models while preserving general‑capability performance.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.

arXiv AI

Jun 16

Localizing Credit at the Divergence: Path-Conditioned Self-Distillation for LLM Reasoning

arXiv:2606. 15576v1 Announce Type: cross Abstract: Reinforcement learning from verifiable rewards assigns a single scalar to each rollout, leaving token-level credit assignment underspecified in long reasoning traces.

By Yu Li, Shu Hong, Tian Lan

llms reinforcement-learning efficiency benchmarks

arXiv AI

Oct 15

PreSO: Privileged Representation Isolation for On-Policy Self-Distillation

arXiv:2610.14001v1 Announce Type: cross Abstract: On-policy self-distillation (OPSD) uses a self-teacher conditioned on privileged information to turn the sparse outcome signal of reinforcement learn...

By Yuhang Yang, Mingxuan Xia, Chao Ye, Shuai Zhu, Shouqing Yang, Haobo Wang, Shenzhi Yang, Zhengqing Zang, Chenglong Wang

reinforcement-learning efficiency benchmarks

Hugging Face Trending Papers

Jun 14

Localizing Credit at the Divergence: Path-Conditioned Self-Distillation for LLM Reasoning

Reinforcement learning from verifiable rewards assigns a single scalar to each rollout, leaving token-level credit assignment underspecified in long reasoning traces. On-policy self-distillation addresses this by letting the same model act as a teacher conditioned on privileged information, producing a dense per-token signal.

llms reinforcement-learning efficiency benchmarks

arXiv Machine Learning

Aug 7

On-Policy Self-Distillation without Any Supervision

arXiv:2608. 06296v1 Announce Type: new Abstract: On-policy (Self-)Distillation (OPD / OPSD) has shown strong potential for post-training large language models (LLMs).

By Yijiang Li, Bingyang Wang, Yijun Liang, Yunjie Tian, Di Fu, Nuno Vasconcelos

llms efficiency benchmarks

Hugging Face Trending Papers

Aug 6

On-Policy Self-Distillation without Any Supervision

On-policy (Self-)Distillation (OPD / OPSD) has shown strong potential for post-training large language models (LLMs). However, existing methods still rely heavily on external supervision, including ground-truth signals, environmental feedback, or guidance from larger models, and therefore fall short of genuine "self"-distillation.

llms efficiency benchmarks

arXiv AI

Sep 30

AdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation

arXiv:2609.38142v1 Announce Type: new Abstract: A small trainable advisor can steer a frozen language-model executor using natural-language advice. In addition to learning from task rewards, the advi...

By Rishabh Agrawal, Hejie Cui, Shasha Li, Shanchan Wu, Sercan \"{O}. Ar{\i}k

llms reinforcement-learning efficiency

One email a morning, machine-written

Related stories

Localizing Credit at the Divergence: Path-Conditioned Self-Distillation for LLM Reasoning

PreSO: Privileged Representation Isolation for On-Policy Self-Distillation

Localizing Credit at the Divergence: Path-Conditioned Self-Distillation for LLM Reasoning

On-Policy Self-Distillation without Any Supervision

On-Policy Self-Distillation without Any Supervision

AdviSD: Learning to Advise Frontier LLMs via Targeted Multi-Turn Self-Distillation