Image de couverture de livre

Harnessing Uncertainty: Entropy-Modulated Policy Gradients for Long-Horizon LLM Agents

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Wang, Jiawei, Liu, Jiacai, Fu, Yuqian, Li, Yingru, Wang, Xintao, Lin, Yuan, Yue, Yu, Zhang, Lin, Wang, Yang, Wang, Ke
Format:	Preprint
Publié:	2025
Sujets:	Machine Learning Computation and Language
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Trust Region Masking for Long-Horizon LLM Reinforcement Learning
par: Li, Yingru, et autres
Publié: (2025)

A Note on Hybrid Online Reinforcement and Imitation Learning for LLMs: Formulations and Algorithms
par: Li, Yingru, et autres
Publié: (2025)

Robust Search with Uncertainty-Aware Value Models for Language Model Reasoning
par: Yu, Fei, et autres
Publié: (2025)

O-Mem: Omni Memory System for Personalized, Long Horizon, Self-Evolving Agents
par: Wang, Piaohong, et autres
Publié: (2025)

Memex(RL): Scaling Long-Horizon LLM Agents via Indexed Experience Memory
par: Wang, Zhenting, et autres
Publié: (2026)

The Optimal Token Baseline: Variance Reduction for Long-Horizon LLM-RL
par: Li, Yingru, et autres
Publié: (2026)

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

Milestone-Guided Policy Learning for Long-Horizon Language Agents
par: Wang, Zixuan, et autres
Publié: (2026)

MAGE: Safeguarding LLM Agents against Long-Horizon Threats via Shadow Memory
par: Wang, Yuhui, et autres
Publié: (2026)

ColorBench: Benchmarking Mobile Agents with Graph-Structured Framework for Complex Long-Horizon Tasks
par: Song, Yuanyi, et autres
Publié: (2025)

VideoSeek: Long-Horizon Video Agent with Tool-Guided Seeking
par: Lin, Jingyang, et autres
Publié: (2026)

Scaling Flaws of Verifier-Guided Search in Mathematical Reasoning
par: Yu, Fei, et autres
Publié: (2025)

OdysseyBench: Evaluating LLM Agents on Long-Horizon Complex Office Application Workflows
par: Wang, Weixuan, et autres
Publié: (2025)

Scaling Long-Horizon LLM Agent via Context-Folding
par: Sun, Weiwei, et autres
Publié: (2025)

EPO: Entropy-regularized Policy Optimization for LLM Agents Reinforcement Learning
par: Xu, Wujiang, et autres
Publié: (2025)

Elementary Analysis of Policy Gradient Methods
par: Liu, Jiacai, et autres
Publié: (2024)

UltraHorizon: Benchmarking Agent Capabilities in Ultra Long-Horizon Scenarios
par: Luo, Haotian, et autres
Publié: (2025)

CHI-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?
par: Chen, Haolin, et autres
Publié: (2026)

ClinEnv: An Interactive Multi-Stage Long Horizon EHR Environment for Agents
par: Lu, Yuxing, et autres
Publié: (2026)

COMAP: Co-Evolving World Models and Agent Policies for LLM Agents
par: Liu, Youwei, et autres
Publié: (2026)

ReFlect: An Effective Harness System for Complex Long-Horizon LLM Reasoning
par: Huang, Fan
Publié: (2026)

PlugMem: A Task-Agnostic Plugin Memory Module for LLM Agents
par: Yang, Ke, et autres
Publié: (2026)

FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents
par: Jia, Haoxuan, et autres
Publié: (2026)

REDSearcher: A Scalable and Cost-Efficient Framework for Long-Horizon Search Agents
par: Chu, Zheng, et autres
Publié: (2026)

Connecting the Dots: Benchmarking Reflective Memory in Long-Horizon Dialogue
par: Lin, Jingjie, et autres
Publié: (2026)

AgentFugue: Agent Scaling for Long-Horizon Tasks through Collective Reasoning
par: Hu, Yuyang, et autres
Publié: (2026)

On the Convergence of Projected Policy Gradient for Any Constant Step Sizes
par: Liu, Jiacai, et autres
Publié: (2023)

LLM$\times$MapReduce-V2: Entropy-Driven Convolutional Test-Time Scaling for Generating Long-Form Articles from Extremely Long Resources
par: Wang, Haoyu, et autres
Publié: (2025)

WildClawBench: A Benchmark for Real-World, Long-Horizon Agent Evaluation
par: Ding, Shuangrui, et autres
Publié: (2026)

Stabilizing Off-Policy Training for Long-Horizon LLM Agent via Turn-Level Importance Sampling and Clipping-Triggered Normalization
par: Li, Chenliang, et autres
Publié: (2025)

Skills on the Fly: Test-Time Adaptive Skill Synthesis for LLM Agents
par: Wang, Jingxing, et autres
Publié: (2026)

HE-SNR: Uncovering Latent Logic via Entropy for Guiding Mid-Training on SWE-bench
par: Wang, Yueyang, et autres
Publié: (2026)

Uncertainty-Aware Variational Reward Factorization via Probabilistic Preference Bases for LLM Personalization
par: Lee, Gyuseok, et autres
Publié: (2026)

EntropyLong: Effective Long-Context Training via Predictive Uncertainty
par: Jia, Junlong, et autres
Publié: (2025)

Harnessing the Unseen: The Hidden Influence of Intrinsic Knowledge in Long-Context Language Models
par: Fu, Yu, et autres
Publié: (2025)

(Perhaps) Beyond Human Translation: Harnessing Multi-Agent Collaboration for Translating Ultra-Long Literary Texts
par: Wu, Minghao, et autres
Publié: (2024)

DictLLM: Harnessing Key-Value Data Structures with Large Language Models for Enhanced Medical Diagnostics
par: Guo, YiQiu, et autres
Publié: (2024)

RAT: Retrieval Augmented Thoughts Elicit Context-Aware Reasoning in Long-Horizon Generation
par: Wang, Zihao, et autres
Publié: (2024)

Why Reasoning Fails to Plan: A Planning-Centric Analysis of Long-Horizon Decision Making in LLM Agents
par: Wang, Zehong, et autres
Publié: (2026)

Dual Hierarchical Dialogue Policy Learning for Legal Inquisitive Conversational Agents
par: Lin, Xubo, et autres
Publié: (2026)