Long-Horizon LLM Agents

LLM: Large Language Model

Long-horizon LLM agents are systems designed to pursue goals through many connected actions, often while using tools and responding to changing information. Common approaches include hierarchical planning and subgoals, reinforcement learning with step-level or milestone rewards, persistent memory and context compression, and harnesses that verify actions and recover from errors. Researchers evaluate them in interactive settings such as ALFWorld, WebShop, ScienceWorld, WebArena, software-engineering benchmarks, and deep-research tasks. A central challenge is making performance reliable over time: errors and lost constraints can compound, while memory, verification, and extra computation can improve success but add cost or introduce new failure modes.

Latest papers

451 papers92 papers in the last four weeks, up 56% on the four weeks before. 0.5% of all new papers.

ExComm: Exploration-Stage Communication for Error-Resilient Agentic Test-Time Scaling
May 21, 2026Woomin Song, Beomjun Kim, Daewon Choi +4Test-Time Scaling Multi-Agent Systems
1KAIST · 2Amazon AGI · 3Together AI
Memory-R2: Fair Credit Assignment for Long-Horizon Memory-Augmented LLM Agents
May 20, 2026Sikuan Yan, Ahmed Bahloul, Ercong Nie +4Group Relative Policy Optimization Credit Assignment in RL
Ludwig Maximilian University of Munich · Munich Center for Machine Learning · Huawei Heisenberg Research Center (Munich) +1
SMDD-Bench: Can LLMs Solve Real-World Small Molecule Drug Design Tasks?
May 20, 2026Kevin Han, Renfei Zhang, Kathy Wei +3LLM Agent Evaluation AI Agent Benchmarks
Carnegie Mellon University · Stealth · Pennsylvania State University
APEX: Autonomous Policy Exploration for Self-Evolving LLM Agents
May 20, 2026Yibo Li, Jiashuo Yang, Zhi Zheng +5LLM Agents Long-Horizon LLM Agents
1National University of Singapore · 2Beijing University of Posts and Telecommunications
MemGym: a Long-Horizon Memory Environment for LLM Agents
May 20, 2026Wujiang Xu, Yu Wang, Kai Mei +8LLM Agent Memory LLM Agent Evaluation
1Rutgers University · 2Capital One · 3Princeton University +1
A Methodology for Selecting and Composing Runtime Architecture Patterns for Production LLM Agents
May 19, 2026Vasundra SrinivasanLLM Agent Orchestration LLM Agents
AI Architect, Independent Researcher · Stanford School of Engineering
STAR-PólyaMath: Multi-Agent Reasoning under Persistent Meta-Strategic Supervision
May 19, 2026Jiaao Wu, Xian Zhang, Hanzhang Liu +3Multi-Agent Orchestration Multi-Agent Reasoning
1Tsinghua University · 2Microsoft Research · 3New York University +1
Code as Agent Harness
May 18, 2026Xuying Ning, Katherine Tieu, Dongqi Fu +39LLM Agent Orchestration Long-Horizon LLM Agents
Latent Action Reparameterization for Efficient Agent Inference
May 18, 2026Wenhao Huang, Qingwen Zeng, Qiyue Chen +11LLM Inference Efficiency Latent Action Learning
Université de Montréal · Mila - Quebec AI Institute · The University of Sydney +7
HINT-SD: Targeted Hindsight Self-Distillation for Long-Horizon Agents
May 18, 2026Woongyeng Yeo, Yumin Choi, Taekyung Ki +1Credit Assignment in RL LLM Agent Training
KAIST
Harnessing LLM Agents with Skill Programs
May 18, 2026Hongjun Liu, Yifei Ming, Shafiq Joty +1LLM Agent Self-Improvement LLM Agent Skill Learning
New York University · Salesforce AI Research
Causal Intervention-Based Memory Selection for Long-Horizon LLM Agents
May 17, 2026Saksham Sahai SrivastavaConversational Memory Long-Term Conversational Memory
School of Computing, University of Georgia, Athens, Georgia, USA.
Episodic-Semantic Memory Architecture for Long-Horizon Scientific Agents
May 17, 2026Nikola MilosevicLLM Agent Memory Long-Horizon LLM Agents
Serbian Institute for Artificial Intelligence Research and Development, Fruskogorska 1, Novi Sad, Serbia · Bayer A.G., Müllerstrasse 178, 13353 Berlin, Germany
RecMem: Recurrence-based Memory Consolidation for Efficient and Effective Long-Running LLM Agents
May 15, 2026Zijie Dai, Shiyuan Deng, Sheng Guan +4Long-Horizon LLM Agents Agent Memory Management
Department of Computer Science and Engineering, The Chinese University of Hong Kong · Huawei Cloud · School of Computer Science, Beijing University of Posts and Telecommunications +2
Harnesses for Inference-Time Alignment over Execution Trajectories
May 15, 2026Boyuan Wang, Bochao Li, Minghan Wang +2LLM Planning Agent Harness Optimization
Southern University of Science and Technology
Toward Natural and Companionable Virtual Agents via Cross-Temporal Emotional Modeling
May 15, 2026Feier Qin, Xiao Li, Yi Zheng +5Long-Horizon LLM Agents Memory-Augmented Agents
Communication University of China BeiJing, China · Microsoft Research Asia BeiJing, China · Institute of Artificial Intelligence, China Telecom BeiJing, China
SaaS-Bench: Can Computer-Use Agents Leverage Real-World SaaS to Solve Professional Workflows?
May 15, 2026Kean Shi, Zihang Li, Tianyi Ma +13Computer-Use Agent Benchmarks Computer-Use Agents
1UniPat AI · 2PKU · 3HKU +2
Remember Your Trace: Memory-Guided Long-Horizon Agentic Framework for Consistent and Hierarchical Repository-Level Code Documentation
May 14, 2026Suyoung Bae, Jaehoon Lee, Changkyu Choi +2Software Engineering Agents LLM Agent Memory
Sungkyunkwan University, South Korea · University of Oslo, Norway
RS-Claw: Progressive Active Tool Exploration via Hierarchical Skill Trees for Remote Sensing Agents
May 13, 2026Liangtian Liu, Zeyuan Wang, Ziyu Li +8Remote Sensing Image Understanding Tool-Using Agents
School of Geosciences and Info-Physics, Central South University, Changsha 410083, China · School of Resources and Environment, University of Electronic Science and Technology of China, Xian 710071, China · School of Earth Sciences and Spatial Information Engineering, Hunan University of Science and Technology, Xiangtan 411201, China +1
MAP: A Map-then-Act Paradigm for Long-Horizon Interactive Agent Reasoning
May 13, 2026Yuxin Liu, Ziang Ye, Yueqing Sun +6LLM Agent Training Long-Horizon LLM Agents
University of Science and Technology of China · Meituan · Institution of Automation, Chinese Academy of Sciences +1
Revisiting DAgger in the Era of LLM-Agents
May 13, 2026Changhao Li, Rushi Qiang, Jiawei Huang +4Software Engineering Agents Dataset Aggregation
1Georgia Institute of Technology · 2ETH Zurich
Executable Agentic Memory for GUI Agent
May 12, 2026Zerui Qin, Sheng Yue, Xingyuan Hua +2GUI Agents LLM Agent Memory
1Tsinghua University, China · 2Sun Yat-sen University, China
PRISM: Pareto-Efficient Retrieval over Intent-Aware Structured Memory for Long-Horizon Agents
May 12, 2026Jingyi Peng, Zhongwei Wan, Weiting Liu +1Graph-Based Retrieval Information Retrieval
Singapore Management University · The Ohio State University · Fudan University
No Action Without a NOD: A Heterogeneous Multi-Agent Architecture for Reliable Service Agents
May 12, 2026Zixu Yang, Hang Zheng, Nan Jiang +5AI Agent Reliability Multi-Agent LLM Systems
1X-LANCE Lab, School of Computer Science, Shanghai Jiao Tong University, Shanghai, China · 2Shanghai Innovation Institution, Shanghai, China · 3Jiangsu Key Lab of Language Computing, Suzhou, China
Hindsight Hint Distillation: Scaffolded Reasoning for SWE Agents from CoT-free Answers
May 12, 2026Shengjie Wang, Guanghe Li, Zonghan Yang +1Software Engineering Agents Long-Horizon LLM Agents
1Tsinghua University
Agent-BRACE: Decoupling Beliefs from Actions in Long-Horizon Tasks via Verbalized State Uncertainty
May 12, 2026Joykirat Singh, Zaid Khan, Archiki Prasad +5Partially Observable RL Bayesian Inference
UNC Chapel Hill · Microsoft Research · The University of Texas at Austin
PIVOT: Bridging Planning and Execution in LLM Agents via Trajectory Refinement
May 11, 2026Tuo Zhang, Alin-Ionut Popa, Yan Xu +2LLM Agent Self-Improvement Long-Horizon LLM Agents
Amazon
The Agent Use of Agent Beings: Agent Cybernetics Is the Missing Science of Foundation Agents
May 11, 2026Xinrun Wang, Chang Yang, He Zhao +2Self-Improving Agents Long-Horizon LLM Agents
Singapore Management University · The Hong Kong Polytechnic University · Nanyang Technological University +2
On Problems of Implicit Context Compression for Software Engineering Agents
May 11, 2026Kirill Gelvan, Igor Slinko, Felix Steinbauer +3Software Engineering Agents Long-Horizon LLM Agents
JetBrains Research · Technical University of Munich, Germany
Continual Harness: Online Adaptation for Self-Improving Foundation Agents
May 11, 2026Seth Karten, Joel Zhang, Tersoo Upaa +5Continual Learning for LLM Agents Game-Playing Agents
Princeton University · ARISE Foundation · Google DeepMind

May 21, 2026J/K move · Enter open · S save

ExComm: Exploration-Stage Communication for Error-Resilient Agentic Test-Time Scaling

Woomin Song, Beomjun Kim, Daewon Choi, Sai Muralidhar Jayanthi+3

1KAIST · 2Amazon AGI · 3Together AI

Test-Time Scaling Multi-Agent Systems Long-Horizon LLM Agents Multi-Agent Communication

Open paper AskPDF arXiv

A common failure mode in long-horizon agentic test-time scaling is error propagation, where factual errors or invalid deductions introduced at intermediate steps persist in the agent's belief state and contaminate later reasoning. Existing test-time scaling methods provide limited control over this process, as they often rely on agents to detect their own mistakes, select among flawed trajectories, or refine solutions only after errors have already shaped the reasoning path. We propose ExComm, a communication protocol for exploration-stage agentic test-time scaling. ExComm is motivated by the empirical observation that the majority of intermediate errors in parallel agentic reasoning produce detectable cross-agent factual conflicts. Leveraging the iterative structure of agentic workflows, ExComm periodically audits agent belief states to detect such conflicts, resolves them through a dedicated tool-based verification loop, and returns concise, targeted feedback to the involved agents. Corrections are incorporated through soft belief updates, which append verified feedback rather than overwriting existing beliefs. Furthermore, to prevent collapsing trajectory diversity due to communication, ExComm further introduces a trajectory diversification module that redirects redundant trajectories toward orthogonal strategies. Experiments on AIME 2024, AIME 2025, and GAIA with Gemini-2.5-Flash-Lite and Qwen3.5-4B show that ExComm consistently outperforms strong test-time scaling baselines, achieving average performance gains of 5.7% and 5.0% over the best-performing baselines, respectively. Further analyses demonstrate improved error recovery, favorable scaling behavior, stronger diversity than adapted communication baselines, and the best performance-cost trade-off among the evaluated methods.