#long-horizon tasks
8 papers · 1 filter
AgentRadio: Passive Awareness for Long-Horizon Multi-Agent Collaboration
Xinxing Ren, Qianbo Zang, Ziyan Wang +4
The paper introduces AgentRadio, an asynchronous message‑passing layer that enables multiple LLM coding agents to stay passively aware of each other's findings during long‑horizon…
RRM: Experience-Driven Reflective Retrieval Memory for Long-Horizon Multimodal Reasoning
Jingxiang Fan, Junbao Zhuo, Bochao Zou
The paper proposes Reflective Retrieval Memory (RRM), a framework that adds a reflective experience memory to an entity‑centric multimodal memory graph, enabling agents to learn an…
Mind the Gap: Promises and Pitfalls of Hierarchical Planning in LeWorldModel
Niccolò Caselli, Francesco Massafra, Samuele Punzo +3
The paper examines adding a high‑level planner to a frozen latent world model (LeWorldModel) for long‑horizon goal‑conditioned control, showing that hierarchical planning can impro…
ABot-AgentOS: A General Robotic Agent OS with Lifelong Multi-modal Memory
Jiayi Tian, Shiao Liu, Yuting Xu +31
The paper introduces ABot-AgentOS, a general operating system layer for robotic agents that adds deliberative planning, multi‑modal memory, verification, and cloud‑edge collaborati…
Diagnosing Semantic Handoff Failures in Agent-Orchestrated Vision-Language-Action Skill Composition
Ke Rui, Yushen Zuo, Jiawei Wang +4
The paper investigates why robots fail when chaining language‑conditioned skills for long‑horizon household tasks, introducing a vision‑language‑action harness that checks skill ch…
ECHO: Prune To Act, Trace To Learn With Selective Turn Memory In Agentic RL
Zijun Xie, Binbin Zheng, Enlei Gong +7
The paper introduces ECHO, a framework that compresses each interaction turn of a language agent into indexed memory records and selectively reconstructs context for decision makin…