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

• URL: http://arxiv.org/abs/2605.18529v1
• Date: Mon, 18 May 2026 15:14:34 GMT
• Title: AMR-SD: Asymmetric Meta-Reflective Self-Distillation for Token-Level Credit Assignment
• Abstract summary: Asymmetric Meta-Reflective Self-Distillation (AMR-SD)<n>We introduce Causal Information Gain (CIG) with an asymmetric, ReLU-gated threshold to translate these reflections into sparse, highly precise token-level advantage modulations.<n> Experiments across scientific, mathematical, and tool-use benchmarks demonstrate that AMR-SD significantly outperforms existing baselines.
• Score: 39.63424981516754
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: 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.

Related papers

• ARISE-RL: Agentic Rubric-Grounded Iterative Self-Evolution with Reinforcement Learning [74.19730275669227]
  ARISE-RL is a novel full-cycle self-evolution framework for training open-ended agents.<n>We present Reward-Gated Self-Evolution Distillation (RG-SED), which distills a memory-augmented variant of the same policy back into itself.<n>We also present ECR-Bench, an expert-calibrated rubric benchmark suite covering single-tool deep research and multi-tool travel planning.
  arXiv  Detail & Related papers  (2026-09-01T10:54:13Z)
• Not All Tokens Deserve Equal Credit: Counterfactual Sensitivity Credit Reallocation for Long-CoT Reasoning [3.9783774144289623]
  Critic-free methods such as GRPO convert response-level rewards into advantages and uniformly broadcast them across tokens.<n>We propose Counterfactual Sensitivity Credit Reallocation (CSCR), a simple extension of GRPO that reduces credit for highly sensitive tokens.<n>On long-CoT mathematical reasoning benchmarks, CSCR consistently outperforms GRPO baseline with the same number of policy updates.
  arXiv  Detail & Related papers  (2026-07-30T09:03:33Z)
• Purified OPSD: On-Policy Self-Distillation Without Losing How to Think [52.879387744920415]
  We find that on-policy self-distillation (OPSD) consistently fails on long chain-of-thought (long-CoT) reasoning models.<n>We propose a two-step solution: first, we construct a reference-only teacher to isolate the non-transferable component of the supervision signal.<n>Second, we use pointwise mutual information (PMI) as the mechanism to transform this residual into a well-formed PMI target distribution.
  arXiv  Detail & Related papers  (2026-07-02T14:33:07Z)
• Rethinking Reward Supervision: Rubric-Conditioned Self-Distillation [60.55792673956761]
  We propose textbfRubric-Conditioned Self-Distillation, a framework that incorporates rubrics as structured, fine-grained feedback for on-policy self-distillation.<n>We evaluate on a diverse suite of science reasoning benchmarks and results show that rubric-conditioned self-distillation effectively converts rubric-level criteria into token-level guidance.
  arXiv  Detail & Related papers  (2026-06-17T17:54:04Z)
• Clipping Bottleneck: Stabilizing RLVR via Stochastic Recovery of Near-Boundary Signals [83.0127582612634]
  Near-boundary Rescue (NSR) is a minimal, plug-and-play modification that retains out-of-bound tokens to recover lost signals.<n>NSR substantially improves training stability and delivers consistent gains over strong baselines such as DAPO and GSPO.
  arXiv  Detail & Related papers  (2026-05-21T16:45:31Z)
• Tailoring Teaching to Aptitude: Direction-Adaptive Self-Distillation for LLM Reasoning [41.384652481442735]
  We propose textbfDirection-Adaptive Self-Distillation (textbfDASD), which reframes privileged self-distillation from uniform teacher imitation into entropy-routed directional supervision.<n>Across six mathematical reasoning benchmarks, DASD achieves the best macro Avg@16 over strong RLVR and self-distillation baselines.
  arXiv  Detail & Related papers  (2026-05-21T10:07:46Z)
• Delayed Homomorphic Reinforcement Learning for Environments with Delayed Feedback [11.866061471514582]
  Delayed feedback breaks the Markov assumption and impedes learning and control.<n>We propose a framework grounded in MDP homomorphisms that collapses belief-equivalent augmented states.<n> Experiments on continuous control tasks in MuJoCo benchmark confirm that our algorithm outperforms strong augmentation-based baselines.
  arXiv  Detail & Related papers  (2026-04-04T08:38:52Z)
• Optimizing RAG Rerankers with LLM Feedback via Reinforcement Learning [47.963236269307735]
  Rerankers play a pivotal role in refining retrieval results for Retrieval-Augmented Generation.<n>Current reranking models are typically optimized on static human annotated relevance labels in isolation, decoupled from the downstream generation process.<n>We introduce ReRanking Preference Optimization (RRPO), a reinforcement learning framework that directly aligns reranking with the LLM's generation quality.
  arXiv  Detail & Related papers  (2026-04-02T14:19:47Z)
• Know What You Know: Metacognitive Entropy Calibration for Verifiable RL Reasoning [31.629261193485053]
  Large reasoning models (LRMs) have emerged as a powerful paradigm for solving complex real-world tasks.<n>Most existing outcome-only RLVR pipelines rely almost exclusively on a binary correctness signal and largely ignore the model's intrinsic uncertainty.<n>We propose EGPO, a metacognitive entropy calibration framework that explicitly integrates intrinsic uncertainty into RLVR for enhancing LRMs.
  arXiv  Detail & Related papers  (2026-02-26T08:40:06Z)
• CARE What Fails: Contrastive Anchored-REflection for Verifiable Multimodal [84.71254539482369]
  Group-relative reinforcement learning with verifiable rewards (RLVR) often wastes the most informative data it already has the failures.<n>We present CARE, a failure-centric post-training framework for multimodal reasoning that turns errors into supervision.<n> CARE improves accuracy and training smoothness while explicitly increasing the share of learning signal that comes from failures.
  arXiv  Detail & Related papers  (2025-12-22T16:34:21Z)
• Retrieval is Not Enough: Enhancing RAG Reasoning through Test-Time Critique and Optimization [58.390885294401066]
  Retrieval-augmented generation (RAG) has become a widely adopted paradigm for enabling knowledge-grounded large language models (LLMs)<n>RAG pipelines often fail to ensure that model reasoning remains consistent with the evidence retrieved, leading to factual inconsistencies or unsupported conclusions.<n>We propose AlignRAG, a novel iterative framework grounded in Critique-Driven Alignment (CDA)<n>We introduce AlignRAG-auto, an autonomous variant that dynamically terminates refinement, removing the need to pre-specify the number of critique iterations.
  arXiv  Detail & Related papers  (2025-04-21T04:56:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.