PulseTrendingJournal ClubResearchersJournalsExplore
Instagram
HomeTrendingJournal ClubExplore
Synapse
⌘+K
Synapse
September 20, 2025

L2M2: A Hierarchical Framework Integrating Large Language Model and Multi-agent Reinforcement Learning

View Full Paper
Ask AI
Bookmark
Share

Authors

MGMinghong GengSPShubham PateriaBSBudhitama Subagdja

Discussion

Loading...

Member takes

Overview

L2M2 demonstrates improved performance in multi-agent tasks with zero-shot planning and without subgoals.

Key Points

  • L2M2's framework enables zero-shot planning, improving multi-agent task handling while reducing training sample usage.
  • In experiments, L2M2's LLM-guided MARL achieved superior performance with under 20% of baseline training samples.
  • Analysis indicates L2M2 can automatically generate navigation plans, enhancing effectiveness in complex scenarios.
  • The framework showcases robust abilities in both subgoal-defined tasks and those lacking predefined objectives.

Cite This Study

Geng et al. (2025) studied this question.

synapsesocial.com/papers/68d46aa631b076d99fa67363https://doi.org/10.24963/ijcai.2025/12
View Full Paper
Ask AI
Bookmark
Share

Also Consider

Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context:

  1. 1A hierarchical multi-agent reinforcement learning framework with high-level guidance from large language models2026 · 1 citations
  2. 2LLM-based Multi-Agent Reinforcement Learning: Current and Future Directions2024 · 12 citations
  3. 3Large Language Model Based Multi-agents: A Survey of Progress and Challenges2024 · 248 citations
  4. 4A Cascading Cooperative Multi-agent Framework for On-ramp Merging Control Integrating Large Language Models2025
  5. 5Hypothetical Minds: Scaffolding Theory of Mind for Multi-Agent Tasks with Large Language Models2024 · 3 citations