Image de couverture de livre

From Generic Correlation to Input-Specific Credit in On-Policy Self Distillation

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Shen, Guobin, Huang, Lei, Cheng, Xiang, Zhao, Chenxiao, Li, Jindong, Zhao, Dongcheng, Yu, Xing
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Anti-Self-Distillation for Reasoning RL via Pointwise Mutual Information
par: Shen, Guobin, et autres
Publié: (2026)

VESPO: Variational Sequence-Level Soft Policy Optimization for Stable Off-Policy LLM Training
par: Shen, Guobin, et autres
Publié: (2026)

Multi-Level Safety Continual Projection for Fine-Tuned Large Language Models without Retraining
par: Han, Bing, et autres
Publié: (2025)

Light Alignment Improves LLM Safety via Model Self-Reflection with a Single Neuron
par: Shen, Sicheng, et autres
Publié: (2026)

Safety Instincts: LLMs Learn to Trust Their Internal Compass for Self-Defense
par: Shen, Guobin, et autres
Publié: (2025)

OISD: On-Policy Internal Self-Distillation of Language Models
par: Liu, Xinyu, et autres
Publié: (2026)

Self-ReSET: Learning to Self-Recover from Unsafe Reasoning Trajectories
par: Zhang, Dongcheng, et autres
Publié: (2026)

PACED: Distillation and On-Policy Self-Distillation at the Frontier of Student Competence
par: Xu, Yuanda, et autres
Publié: (2026)

Online Policy Distillation with Decision-Attention
par: Yu, Xinqiang, et autres
Publié: (2024)

HDPO: Hybrid Distillation Policy Optimization via Privileged Self-Distillation
par: Ding, Ken
Publié: (2026)

A General Framework for Learning from Weak Supervision
par: Chen, Hao, et autres
Publié: (2024)

OPD+: Rethinking the Advantage Design for On-Policy Distillation
par: Zhao, Hanyang, et autres
Publié: (2026)

Towards Reliable Evaluation of Adversarial Robustness for Spiking Neural Networks
par: Wang, Jihang, et autres
Publié: (2025)

On-Policy Distillation of Language Models: Learning from Self-Generated Mistakes
par: Agarwal, Rishabh, et autres
Publié: (2023)

Beyond Uniform Credit: Causal Credit Assignment for Policy Optimization
par: Khandoga, Mykola, et autres
Publié: (2026)

Implicit Federated In-context Learning For Task-Specific LLM Fine-Tuning
par: Li, Dongcheng, et autres
Publié: (2025)

Graph Dimension Attention Networks for Enterprise Credit Assessment
par: Wei, Shaopeng, et autres
Publié: (2024)

Multimodal Generative Engine Optimization: Rank Manipulation for Vision-Language Model Rankers
par: Du, Yixuan, et autres
Publié: (2026)

Self-Supervised Quantization-Aware Knowledge Distillation
par: Zhao, Kaiqi, et autres
Publié: (2024)

FedSDR: Federated Self-Distillation with Rectification
par: Ren, Ziheng, et autres
Publié: (2026)

Bootstrapping Exploration with Group-Level Natural Language Feedback in Reinforcement Learning
par: Huang, Lei, et autres
Publié: (2026)

Proximal Policy Distillation
par: Spigler, Giacomo
Publié: (2024)

When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning
par: Liu, Xiaogeng, et autres
Publié: (2026)

Unifying Group-Relative and Self-Distillation Policy Optimization via Sample Routing
par: Li, Gengsheng, et autres
Publié: (2026)

TRACE: Distilling Where It Matters via Token-Routed Self On-Policy Alignment
par: Wang, Jiaxuan, et autres
Publié: (2026)

Revisiting On-Policy Distillation: Empirical Failure Modes and Simple Fixes
par: Fu, Yuqian, et autres
Publié: (2026)

Asymmetric On-Policy Distillation: Bridging Exploitation and Imitation at the Token Level
par: Jia, Nan, et autres
Publié: (2026)

PromptBench: A Unified Library for Evaluation of Large Language Models
par: Zhu, Kaijie, et autres
Publié: (2023)

Dynamic Evaluation of Large Language Models by Meta Probing Agents
par: Zhu, Kaijie, et autres
Publié: (2024)

Prompt Tuning with Diffusion for Few-Shot Pre-trained Policy Generalization
par: Hu, Shengchao, et autres
Publié: (2024)

DGPO: Distribution Guided Policy Optimization for Fine Grained Credit Assignment
par: Jin, Hongbo, et autres
Publié: (2026)

Generalized Policy Gradient with History-Aware Decision Transformer for Reliable Routing over Graph Signals
par: Wei, Xing, et autres
Publié: (2025)

BPL: Bias-adaptive Preference Distillation Learning for Recommender System
par: Kang, SeongKu, et autres
Publié: (2025)

OGLS-SD: On-Policy Self-Distillation with Outcome-Guided Logit Steering for LLM Reasoning
par: Yang, Yuxiao, et autres
Publié: (2026)

Provable and Practical In-Context Policy Optimization for Self-Improvement
par: Yu, Tianrun, et autres
Publié: (2026)

Annealing Self-Distillation Rectification Improves Adversarial Training
par: Wu, Yu-Yu, et autres
Publié: (2023)

Model-Based Proactive Cost Generation for Learning Safe Policies Offline with Limited Violation Data
par: Xue, Ruiqi, et autres
Publié: (2026)

Extreme Region Policy Distillation
par: Chen, Changyu, et autres
Publié: (2026)

Towards Better Generalization via Distributional Input Projection Network
par: Hao, Yifan, et autres
Publié: (2025)

Self-Distillation for Multi-Token Prediction
par: Zhao, Guoliang, et autres
Publié: (2026)