Imagen de Portada

Mitigating Think-Answer Mismatch in LLM Reasoning Through Noise-Aware Advantage Reweighting

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Shen, Si, Shen, Peijun, Zhao, Wenhua, Zhu, Danhao
Formato:	Preprint
Publicado:	2025
Materias:	Machine Learning
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

RevOrder: A Novel Method for Enhanced Arithmetic in Language Models
por: Shen, Si, et al.
Publicado: (2024)

LLM-Metrics: Measuring Research Impact Through Large Language Model Memory
por: Shen, Si, et al.
Publicado: (2026)

CurveRL: Principled Distribution-Aware Context Reweighting for LLM Reasoning
por: Sun, Ke, et al.
Publicado: (2026)

GEAR: Granularity-Adaptive Advantage Reweighting for LLM Agents via Self-Distillation
por: Li, Sijia, et al.
Publicado: (2026)

TIAR: Trajectory-Informed Advantage Reweighting for LLM Abstention Learning
por: Pan, Muyu, et al.
Publicado: (2026)

Taming Extreme Tokens: Covariance-Aware GRPO with Gaussian-Kernel Advantage Reweighting
por: Wang, Cheng, et al.
Publicado: (2026)

Uncertainty-Aware Answer Selection for Improved Reasoning in Multi-LLM Systems
por: Agrawal, Aakriti, et al.
Publicado: (2025)

Kernelized Advantage Estimation: From Nonparametric Statistics to LLM Reasoning
por: Gong, Shijin, et al.
Publicado: (2026)

Learn to Think: Bootstrapping LLM Reasoning Capability Through Graph Representation Learning
por: Gao, Hang, et al.
Publicado: (2025)

Think Dense, Not Long: Dynamic Decoupled Conditional Advantage for Efficient Reasoning
por: Peng, Keqin, et al.
Publicado: (2026)

Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization
por: Zhang, Qingyang, et al.
Publicado: (2025)

Thinking with Knowledge Graphs: Enhancing LLM Reasoning Through Structured Data
por: Wu, Xue, et al.
Publicado: (2024)

Policy Filtration for RLHF to Mitigate Noise in Reward Models
por: Zhang, Chuheng, et al.
Publicado: (2024)

Efficient Reasoning with Hidden Thinking
por: Shen, Xuan, et al.
Publicado: (2025)

Topology-Aware Dynamic Reweighting for Distribution Shifts on Graph
por: Zheng, Weihuang, et al.
Publicado: (2024)

Mitigating the Noise Shift for Denoising Generative Models via Noise Awareness Guidance
por: Zhong, Jincheng, et al.
Publicado: (2025)

Accelerating RL for LLM Reasoning with Optimal Advantage Regression
por: Brantley, Kianté, et al.
Publicado: (2025)

Quantile Advantage Estimation: Stabilizing RLVR for LLM Reasoning
por: Wu, Junkang, et al.
Publicado: (2025)

Flow-of-Options: Diversified and Improved LLM Reasoning by Thinking Through Options
por: Nair, Lakshmi, et al.
Publicado: (2025)

Individual Fairness Through Reweighting and Tuning
por: Mahamadou, Abdoul Jalil Djiberou, et al.
Publicado: (2024)

Stable Adaptive Thinking via Advantage Shaping and Length-Aware Gradient Regulation
por: Xu, Zihang, et al.
Publicado: (2026)

ThinkEdit: Interpretable Weight Editing to Mitigate Overly Short Thinking in Reasoning Models
por: Sun, Chung-En, et al.
Publicado: (2025)

Arbitrage: Efficient Reasoning via Advantage-Aware Speculation
por: Maheswaran, Monishwaran, et al.
Publicado: (2025)

Mitigating the Language Mismatch and Repetition Issues in LLM-based Machine Translation via Model Editing
por: Wang, Weichuan, et al.
Publicado: (2024)

Mitigating Individual Skin Tone Bias in Skin Lesion Classification through Distribution-Aware Reweighting
por: Paxton, Kuniko, et al.
Publicado: (2025)

Thinking Short and Right Over Thinking Long: Serving LLM Reasoning Efficiently and Accurately
por: Wang, Yuhang, et al.
Publicado: (2025)

Exploring Criteria of Loss Reweighting to Enhance LLM Unlearning
por: Yang, Puning, et al.
Publicado: (2025)

Enhancing Adversarial Training via Reweighting Optimization Trajectory
por: Huang, Tianjin, et al.
Publicado: (2023)

Beyond the Last Answer: Your Reasoning Trace Uncovers More than You Think
por: Hammoud, Hasan Abed Al Kader, et al.
Publicado: (2025)

Mitigating Mismatch within Reference-based Preference Optimization
por: Yuan, Suqin, et al.
Publicado: (2026)

BASIS: Batchwise Advantage Estimation from Single-Rollout Information Sharing for LLM Reasoning
por: Gong, Shijin, et al.
Publicado: (2026)

Think-Augmented Function Calling: Improving LLM Parameter Accuracy Through Embedded Reasoning
por: Wei, Lei, et al.
Publicado: (2026)

Vulnerability-Aware Alignment: Mitigating Uneven Forgetting in Harmful Fine-Tuning
por: Chen, Liang, et al.
Publicado: (2025)

LAD: Learning Advantage Distribution for Reasoning
por: Li, Wendi, et al.
Publicado: (2026)

Classical Verification of Quantum Learning Advantages with Noises
por: Ma, Yinghao, et al.
Publicado: (2024)

Mitigating the Backdoor Effect for Multi-Task Model Merging via Safety-Aware Subspace
por: Yang, Jinluan, et al.
Publicado: (2024)

DAJ: Data-Reweighted LLM Judge for Test-Time Scaling in Code Generation
por: Qin, Peijia, et al.
Publicado: (2026)

Group & Reweight: A Novel Cost-Sensitive Approach to Mitigating Class Imbalance in Network Traffic Classification
por: Du, Wumei, et al.
Publicado: (2024)

DFedReweighting: A Unified Framework for Objective-Oriented Reweighting in Decentralized Federated Learning
por: Zhang, Kaichuang, et al.
Publicado: (2025)

Differential Smoothing Mitigates Sharpening and Improves LLM Reasoning
por: Gai, Jingchu, et al.
Publicado: (2025)