Cover Image

Interpreting and Controlling LLM Reasoning through Integrated Policy Gradient

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Li, Changming, Zhang, Kaixing, Xu, Haoyun, Shi, Yingdong, Zhang, Zheng, Song, Kaitao, Ren, Kan
Format:	Preprint
Published:	2026
Subjects:	Artificial Intelligence Computation and Language Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning
by: Zhang, Yifan, et al.
Published: (2025)

CAPO: Towards Enhancing LLM Reasoning through Generative Credit Assignment
by: Xie, Guofu, et al.
Published: (2025)

Slow-Fast Policy Optimization: Reposition-Before-Update for LLM Reasoning
by: Wang, Ziyan, et al.
Published: (2025)

Klear-Reasoner: Advancing Reasoning Capability via Gradient-Preserving Clipping Policy Optimization
by: Su, Zhenpeng, et al.
Published: (2025)

Seek in the Dark: Reasoning via Test-Time Instance-Level Policy Gradient in Latent Space
by: Li, Hengli, et al.
Published: (2025)

Simple Policy Gradients for Reasoning with Diffusion Language Models
by: Zhan, Anthony
Published: (2025)

Thought Branches: Interpreting LLM Reasoning Requires Resampling
by: Macar, Uzay, et al.
Published: (2025)

MolReasoner: Toward Effective and Interpretable Reasoning for Molecular LLMs
by: Zhao, Guojiang, et al.
Published: (2025)

C$^2$GSPG: Confidence-calibrated Group Sequence Policy Gradient towards Self-aware Reasoning
by: Liu, Haotian, et al.
Published: (2025)

Scaf-GRPO: Scaffolded Group Relative Policy Optimization for Enhancing LLM Reasoning
by: Zhang, Xichen, et al.
Published: (2025)

Internalizing LLM Reasoning via Discovery and Replay of Latent Actions
by: Shi, Zhenning, et al.
Published: (2026)

SABER: Switchable and Balanced Training for Efficient LLM Reasoning
by: Zhao, Kai, et al.
Published: (2025)

Efficient Contextual LLM Cascades through Budget-Constrained Policy Learning
by: Zhang, Xuechen, et al.
Published: (2024)

Advancing General-Purpose Reasoning Models with Modular Gradient Surgery
by: Cai, Min, et al.
Published: (2026)

PORTool: Importance-Aware Policy Optimization with Rewarded Tree for Multi-Tool-Integrated Reasoning
by: Wu, Feijie, et al.
Published: (2025)

Multimodal Fusion of EHR in Structures and Semantics: Integrating Clinical Records and Notes with Hypergraph and LLM
by: Cui, Hejie, et al.
Published: (2024)

AIOS Compiler: LLM as Interpreter for Natural Language Programming and Flow Programming of AI Agents
by: Xu, Shuyuan, et al.
Published: (2024)

SFT-then-RL Outperforms Mixed-Policy Methods for LLM Reasoning
by: Limozin, Alexis, et al.
Published: (2026)

Incentivizing Agentic Reasoning in LLM Judges via Tool-Integrated Reinforcement Learning
by: Xu, Ran, et al.
Published: (2025)

Gradient-Adaptive Policy Optimization: Towards Multi-Objective Alignment of Large Language Models
by: Li, Chengao, et al.
Published: (2025)

Prismatic Synthesis: Gradient-based Data Diversification Boosts Generalization in LLM Reasoning
by: Jung, Jaehun, et al.
Published: (2025)

From Eigenmodes to Proofs: Integrating Graph Spectral Operators with Symbolic Interpretable Reasoning
by: Kiruluta, Andrew, et al.
Published: (2025)

Towards Effective Code-Integrated Reasoning
by: Bai, Fei, et al.
Published: (2025)

Dual-Uncertainty Guided Policy Learning for Multimodal Reasoning
by: Liu, Rui, et al.
Published: (2025)

How Is LLM Reasoning Distracted by Irrelevant Context? An Analysis Using a Controlled Benchmark
by: Yang, Minglai, et al.
Published: (2025)

Reinforce LLM Reasoning through Multi-Agent Reflection
by: Yuan, Yurun, et al.
Published: (2025)

GTPO: Stabilizing Group Relative Policy Optimization via Gradient and Entropy Control
by: Simoni, Marco, et al.
Published: (2025)

Exploring LLM Reasoning Through Controlled Prompt Variations
by: Chatziveroglou, Giannis, et al.
Published: (2025)

Large Language Models Explore by Latent Distilling
by: Zeng, Yuanhao, et al.
Published: (2026)

Optimizing Chain-of-Thought Reasoners via Gradient Variance Minimization in Rejection Sampling and RL
by: Yao, Jiarui, et al.
Published: (2025)

KORGym: A Dynamic Game Platform for LLM Reasoning Evaluation
by: Shi, Jiajun, et al.
Published: (2025)

GPG: Generalized Policy Gradient Theorem for Transformer-based Policies
by: Mao, Hangyu, et al.
Published: (2025)

Effectively Controlling Reasoning Models through Thinking Intervention
by: Wu, Tong, et al.
Published: (2025)

Learning to Reason under Off-Policy Guidance
by: Yan, Jianhao, et al.
Published: (2025)

FlowRL: Matching Reward Distributions for LLM Reasoning
by: Zhu, Xuekai, et al.
Published: (2025)

Empowering Multi-Turn Tool-Integrated Agentic Reasoning with Group Turn Policy Optimization
by: Ding, Yifeng, et al.
Published: (2025)

Better LLM Reasoning via Dual-Play
by: Zhang, Zhengxin, et al.
Published: (2025)

Formal-LLM: Integrating Formal Language and Natural Language for Controllable LLM-based Agents
by: Li, Zelong, et al.
Published: (2024)

Crosslingual Reasoning through Test-Time Scaling
by: Yong, Zheng-Xin, et al.
Published: (2025)

Reasoning Is Not Free: Robust Adaptive Cost-Efficient Routing for LLM-as-a-Judge
by: Zhang, Wenbo, et al.
Published: (2026)