Imagen de Portada

KTAE: A Model-Free Algorithm to Key-Tokens Advantage Estimation in Mathematical Reasoning

Fuente: arXiv

Guardado en:

Detalles Bibliográficos
Autores principales:	Sun, Wei, Yang, Wen, Jian, Pu, Du, Qianlong, Cui, Fuwei, Ren, Shuo, Zhang, Jiajun
Formato:	Preprint
Publicado:	2025
Materias:	Artificial Intelligence Computation and Language
Acceso en línea:	Acceder al recurso
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

Ejemplares similares

An Efficient and Precise Training Data Construction Framework for Process-supervised Reward Model in Mathematical Reasoning
por: Sun, Wei, et al.
Publicado: (2025)

Conditional Advantage Estimation for Reinforcement Learning in Large Reasoning Models
por: Chen, Guanxu, et al.
Publicado: (2025)

Key-Point-Driven Mathematical Reasoning Distillation of Large Language Model
por: Zhu, Xunyu, et al.
Publicado: (2024)

From Next-Token to Mathematics: The Learning Dynamics of Mathematical Reasoning in Language Models
por: Mishra, Shubhra, et al.
Publicado: (2024)

ChineseWebText 2.0: Large-Scale High-quality Chinese Web Text with Multi-dimensional and fine-grained information
por: Zhang, Wanyue, et al.
Publicado: (2024)

Teaching Vision-Language Models to Ask: Resolving Ambiguity in Visual Questions
por: Jian, Pu, et al.
Publicado: (2025)

HeartLLM: Discretized ECG Tokenization for LLM-Based Diagnostic Reasoning
por: Yang, Jinning, et al.
Publicado: (2025)

A Survey on Large Language Models for Mathematical Reasoning
por: Wang, Peng-Yuan, et al.
Publicado: (2025)

Markov Chain of Thought for Efficient Mathematical Reasoning
por: Yang, Wen, et al.
Publicado: (2024)

Unmasking Reasoning Processes: A Process-aware Benchmark for Evaluating Structural Mathematical Reasoning in LLMs
por: Zheng, Xiang, et al.
Publicado: (2026)

Distilling Mathematical Reasoning Capabilities into Small Language Models
por: Zhu, Xunyu, et al.
Publicado: (2024)

Step Potential Advantage Estimation: Harnessing Intermediate Confidence and Correctness for Efficient Mathematical Reasoning
por: Wu, Fei, et al.
Publicado: (2026)

Key-Point-Driven Data Synthesis with its Enhancement on Mathematical Reasoning
por: Huang, Yiming, et al.
Publicado: (2024)

Look Again, Think Slowly: Enhancing Visual Reflection in Vision-Language Models
por: Jian, Pu, et al.
Publicado: (2025)

Can Large Models Teach Student Models to Solve Mathematical Problems Like Human Beings? A Reasoning Distillation Method via Multi-LoRA Interaction
por: Li, Xinhe, et al.
Publicado: (2025)

From Implicit to Explicit: Token-Efficient Logical Supervision for Mathematical Reasoning in LLMs
por: Wang, Shaojie, et al.
Publicado: (2026)

PromptCoT: Synthesizing Olympiad-level Problems for Mathematical Reasoning in Large Language Models
por: Zhao, Xueliang, et al.
Publicado: (2025)

A Prompt-Based Knowledge Graph Foundation Model for Universal In-Context Reasoning
por: Cui, Yuanning, et al.
Publicado: (2024)

DetermLR: Augmenting LLM-based Logical Reasoning from Indeterminacy to Determinacy
por: Sun, Hongda, et al.
Publicado: (2023)

Token-Supervised Value Models for Enhancing Mathematical Problem-Solving Capabilities of Large Language Models
por: Lee, Jung Hyun, et al.
Publicado: (2024)

Critical Tokens Matter: Token-Level Contrastive Estimation Enhances LLM's Reasoning Capability
por: Lin, Zicheng, et al.
Publicado: (2024)

Improving Mathematical Reasoning Capabilities of Small Language Models via Feedback-Driven Distillation
por: Zhu, Xunyu, et al.
Publicado: (2024)

SHAPE: Stage-aware Hierarchical Advantage via Potential Estimation for LLM Reasoning
por: Ai, Zhengyang, et al.
Publicado: (2026)

MuMath-Code: Combining Tool-Use Large Language Models with Multi-perspective Data Augmentation for Mathematical Reasoning
por: Yin, Shuo, et al.
Publicado: (2024)

AIMO-2 Winning Solution: Building State-of-the-Art Mathematical Reasoning Models with OpenMathReasoning dataset
por: Moshkov, Ivan, et al.
Publicado: (2025)

Examining False Positives under Inference Scaling for Mathematical Reasoning
por: Wang, Yu, et al.
Publicado: (2025)

MathFimer: Enhancing Mathematical Reasoning by Expanding Reasoning Steps through Fill-in-the-Middle Task
por: Yan, Yuchen, et al.
Publicado: (2025)

WizardMath: Empowering Mathematical Reasoning for Large Language Models via Reinforced Evol-Instruct
por: Luo, Haipeng, et al.
Publicado: (2023)

TokUR: Token-Level Uncertainty Estimation for Large Language Model Reasoning
por: Zhang, Tunyu, et al.
Publicado: (2025)

Parallel Scaling Law: Unveiling Reasoning Generalization through A Cross-Linguistic Perspective
por: Yang, Wen, et al.
Publicado: (2025)

Dynamic Thinking-Token Selection for Efficient Reasoning in Large Reasoning Models
por: Guo, Zhenyuan, et al.
Publicado: (2026)

Speculative Prefill: Turbocharging TTFT with Lightweight and Training-Free Token Importance Estimation
por: Liu, Jingyu, et al.
Publicado: (2025)

Guiding Language Model Reasoning with Planning Tokens
por: Wang, Xinyi, et al.
Publicado: (2023)

Strategy Executability in Mathematical Reasoning: Leveraging Human-Model Differences for Effective Guidance
por: Liang, Weida, et al.
Publicado: (2026)

Understanding and Mitigating Spurious Signal Amplification in Test-Time Reinforcement Learning for Math Reasoning
por: Yu, Yongcan, et al.
Publicado: (2026)

Demystifying Reasoning Dynamics with Mutual Information: Thinking Tokens are Information Peaks in LLM Reasoning
por: Qian, Chen, et al.
Publicado: (2025)

What Really Improves Mathematical Reasoning: Structured Reasoning Signals Beyond Pure Code
por: Zhao, Yuze, et al.
Publicado: (2026)

S^3cMath: Spontaneous Step-level Self-correction Makes Large Language Models Better Mathematical Reasoners
por: Yan, Yuchen, et al.
Publicado: (2024)

Mid-Think: Training-Free Intermediate-Budget Reasoning via Token-Level Triggers
por: Yang, Wang, et al.
Publicado: (2026)

A Survey on Feedback-based Multi-step Reasoning for Large Language Models on Mathematics
por: Wei, Ting-Ruen, et al.
Publicado: (2025)