Image de couverture de livre

Taming Extreme Tokens: Covariance-Aware GRPO with Gaussian-Kernel Advantage Reweighting

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Wang, Cheng, Liu, Qin, Zhou, Wenxuan, Chen, Muhao
Format:	Preprint
Publié:	2026
Sujets:	Computation and Language Artificial Intelligence Machine Learning
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

MMR-GRPO: Accelerating GRPO-Style Training through Diversity-Aware Reward Reweighting
par: Wei, Kangda, et autres
Publié: (2026)

GEAR: Granularity-Adaptive Advantage Reweighting for LLM Agents via Self-Distillation
par: Li, Sijia, et autres
Publié: (2026)

Familiarity-Aware Evidence Compression for Retrieval-Augmented Generation
par: Jung, Dongwon, et autres
Publié: (2024)

MetaScale: Test-Time Scaling with Evolving Meta-Thoughts
par: Liu, Qin, et autres
Publié: (2025)

OmniStruct: Universal Text-to-Structure Generation across Diverse Schemas
par: Huang, James Y., et autres
Publié: (2025)

ThinkGuard: Deliberative Slow Thinking Leads to Cautious Guardrails
par: Wen, Xiaofei, et autres
Publié: (2025)

GRPO-CARE: Consistency-Aware Reinforcement Learning for Multimodal Reasoning
par: Chen, Yi, et autres
Publié: (2025)

Offset Unlearning for Large Language Models
par: Huang, James Y., et autres
Publié: (2024)

Training-Trajectory-Aware Token Selection
par: Shen, Zhanming, et autres
Publié: (2026)

ExGRPO: Learning to Reason from Experience
par: Zhan, Runzhe, et autres
Publié: (2025)

T-REG: Preference Optimization with Token-Level Reward Regularization
par: Zhou, Wenxuan, et autres
Publié: (2024)

From Shortcuts to Triggers: Backdoor Defense with Denoised PoE
par: Liu, Qin, et autres
Publié: (2023)

AgentKernelArena: Generalization-Aware Benchmarking of GPU Kernel Optimization Agents
par: Younesian, Sharareh, et autres
Publié: (2026)

Contrastive Instruction Tuning
par: Yan, Tianyi Lorena, et autres
Publié: (2024)

ArenaBencher: Automatic Benchmark Evolution via Multi-Model Competitive Evaluation
par: Liu, Qin, et autres
Publié: (2025)

Can GRPO Help LLMs Transcend Their Pretraining Origin?
par: Ni, Kangqi, et autres
Publié: (2025)

Mitigating Selection Bias in Large Language Models via Permutation-Aware GRPO
par: Zheng, Jinquan, et autres
Publié: (2026)

GanitLLM: Difficulty-Aware Bengali Mathematical Reasoning through Curriculum-GRPO
par: Dipta, Shubhashis Roy, et autres
Publié: (2026)

Securing Multi-turn Conversational Language Models From Distributed Backdoor Triggers
par: Tong, Terry, et autres
Publié: (2024)

Adaptive-Boundary-Clipping GRPO: Ensuring Bounded Ratios for Stable and Generalizable Training
par: Liu, Chi, et autres
Publié: (2026)

Token-Budget-Aware LLM Reasoning
par: Han, Tingxu, et autres
Publié: (2024)

DACA-GRPO: Denoising-Aware Credit Assignment for Reinforcement Learning in Diffusion Language Models
par: Monsefi, Amin Karimi, et autres
Publié: (2026)

Arbitrage: Efficient Reasoning via Advantage-Aware Speculation
par: Maheswaran, Monishwaran, et autres
Publié: (2025)

Stable Adaptive Thinking via Advantage Shaping and Length-Aware Gradient Regulation
par: Xu, Zihang, et autres
Publié: (2026)

More is not always better? Enhancing Many-Shot In-Context Learning with Differentiated and Reweighting Objectives
par: Zhang, Xiaoqing, et autres
Publié: (2025)

DoGE: Domain Reweighting with Generalization Estimation
par: Fan, Simin, et autres
Publié: (2023)

TIAR: Trajectory-Informed Advantage Reweighting for LLM Abstention Learning
par: Pan, Muyu, et autres
Publié: (2026)

Stepwise Guided Policy Optimization: Coloring your Incorrect Reasoning in GRPO
par: Chen, Peter, et autres
Publié: (2025)

Active Layer-Contrastive Decoding Reduces Hallucination in Large Language Model Generation
par: Zhang, Hongxiang, et autres
Publié: (2025)

Balanced Aggregation: Understanding and Fixing Aggregation Bias in GRPO
par: Zeng, Zhiyuan, et autres
Publié: (2026)

Semantic-Clipping: Efficient Vision-Language Modeling with Semantic-Guidedd Visual Selection
par: Li, Bangzheng, et autres
Publié: (2025)

Filter-then-Weight: Online Data Selection and Reweighting for LLM Fine-Tuning
par: Wang, Fangxin, et autres
Publié: (2026)

mDPO: Conditional Preference Optimization for Multimodal Large Language Models
par: Wang, Fei, et autres
Publié: (2024)

Aligning Frozen LLMs by Reinforcement Learning: An Iterative Reweight-then-Optimize Approach
par: Zhang, Xinnan, et autres
Publié: (2025)

Asymmetric Advantage Modulation Calibrates Entropy Dynamics in RLVR
par: Gu, Hengrui, et autres
Publié: (2026)

Multi-Task GRPO: Reliable LLM Reasoning Across Tasks
par: Ramesh, Shyam Sundhar, et autres
Publié: (2026)

LayerIF: Estimating Layer Quality for Large Language Models using Influence Functions
par: Askari, Hadi, et autres
Publié: (2025)

Stratified GRPO: Handling Structural Heterogeneity in Reinforcement Learning of LLM Search Agents
par: Zhu, Mingkang, et autres
Publié: (2025)

Be My Eyes: Extending Large Language Models to New Modalities Through Multi-Agent Collaboration
par: Huang, James Y., et autres
Publié: (2025)

Less Diverse, Less Safe: The Indirect But Pervasive Risk of Test-Time Scaling in Large Language Models
par: Nahin, Shahriar Kabir, et autres
Publié: (2025)