paper-with-me

홈 › Papers

AMR-SD: Asymmetric Meta-Reflective Self-Distillation for Token-Level Credit Assignment

2026-05-18 · Zhenlin Wei, Pu Jian, Yingzhuo Deng, Xiaohan Wang, Jiajun Chai, Zhexin Hu, Wei Lin, Shanbin Zhang, Guojun Yin arxiv

The alignment of Large Language Models (LLMs) for complex reasoning heavily relies on Reinforcement Learning with Verifiable Rewards (RLVR). However, standard algorithms like GRPO apply sequence-level rewards uniformly to all tokens, creating a severe credit-assignment bottleneck. While on-policy self-distillation attempts to resolve this by conditioning a self-teacher on privileged contexts, direct exposure to raw oracle solutions often induces over-conditioned teacher distributions, implicit answer leakage, and late-stage training collapse. To overcome these limitations, we propose Asymmetric Meta-Reflective Self-Distillation (AMR-SD). Instead of conditioning directly on raw reference traces, AMR-SD inserts a reflection bottleneck: it compresses diagnostic signals -- from verifier outcomes, peer rollouts, or reference feedback -- into concise, self-generated Socratic hints and critiques. Furthermore, we introduce Causal Information Gain (CIG) with an asymmetric, ReLU-gated threshold to translate these reflections into sparse, highly precise token-level advantage modulations. Combined with temporal annealing, this mechanism preserves the base environmental reward while filtering out distributional noise. Experiments across scientific, mathematical, and tool-use benchmarks demonstrate that AMR-SD significantly outperforms existing baselines, achieving robust long-horizon stability and successfully preventing late-stage collapse.

📄 PDF Abstract BibTeX arXiv:2605.18529

Code (0)

등록된 구현이 없습니다.

Tasks

Reinforcement Learning

Similar Papers 제목 키워드 기반

ROSD: Reflective On-Policy Self-Distillation for Language Model Reasoning across Domains

2026-05-27 · Ziqi Zhao, Xinyu Ma, Liu Yang, Yujie Feng 외 arxiv

On-policy self-distillation (OPSD) improves the reasoning performance of large language models (LLMs) by providing dense token-level supervision for on-policy rollouts. However, existing OPSD methods often yield limited …

Domain Generalization

TTPO: Test-Time Policy Optimization

2026-08-27 · Aozhe Wang, Zhengxi Lu, Jianze Wang, Shangke Lv 외 hf

Recent prominent post-training methods, such as Reinforcement Learning (RL) and On-Policy Self-Distillation (OPSD), have driven rapid progress in mathematical reasoning for large language models, yet their reliance on gr…

Mathematical ReasoningReinforcement Learning

3D Distillation: Improving Self-Supervised Monocular Depth Estimation on Reflective Surfaces

2023-01-01 · ICCV 2023 1 · Xuepeng Shi, Georgi Dikov, Gerhard Reitmayr, Tae-Kyun Kim 외

Self-supervised monocular depth estimation (SSMDE) aims at predicting the dense depth maps of monocular images, by learning to minimize a photometric loss using spatially neighboring image pairs during training. Whil…

Depth EstimationMonocular Depth Estimation

Visual Contrastive Self-Distillation

2026-07-23 · Yijun Liang, Yunjie Tian, Yijiang Li, Yuqi Jia 외 arxiv

On-policy self-distillation (OPSD) is promising as it removes the external teacher required by on-policy distillation (OPD), yet it still needs asymmetric information between teacher and student to ensure that the self-t…

Test-Time Scaling with Reflective Generative Model

2025-07-02 · Zixiao Wang, Yuxin Wang, Xiaorui Wang, Mengting Xing 외 arxiv

We introduce our first reflective generative model MetaStone-S1, which obtains OpenAI o3-mini's performance via the new Reflective Generative Form. The new form focuses on high-quality reasoning trajectory selection and …