mlsanityData to 14 Oct 2026

LERO: LLM-driven Evolutionary framework with Hybrid Rewards and Enhanced Observation for Multi-Agent Reinforcement Learning

arXiv 2503.21807 ↗First posted 25 Mar 2025cs.LG cs.AI cs.MAcited by 9

On the map: LLM agents and tool use → Reinforcement learning for LLM agents

Abstract (arXiv)

Multi-agent reinforcement learning (MARL) faces two critical bottlenecks distinct from single-agent RL: credit assignment in cooperative tasks and partial observability of environmental states. We propose LERO, a framework integrating Large language models (LLMs) with evolutionary optimization to address these MARL-specific challenges. The solution centers on two LLM-generated components: a hybrid reward function that dynamically allocates individual credit through reward decomposition, and an observation enhancement function that augments partial observations with inferred environmental context. An evolutionary algorithm optimizes these components through iterative MARL training cycles, where top-performing candidates guide subsequent LLM generations. Evaluations in Multi-Agent Particle Environments (MPE) demonstrate LERO's superiority over baseline methods, with improved task performance and training efficiency.

Key facts. Cited by 9 papers in the map (data through 14 Oct 2026). Placed in Reinforcement learning for LLM agents, one of 16 directions in LLM agents and tool use. Builds on 14 of its 35 references that are in the map. First posted on arXiv 25 Mar 2025.

Builds on

14 of 35 references.

References · 14

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,744
  2. 2017Multi-Agent Actor-Critic for Mixed Cooperative-Competitive Environmentscited by 1,558
  3. 2021The Surprising Effectiveness of PPO in Cooperative, Multi-Agent Gamescited by 942
  4. 2023Eureka: Human-Level Reward Design via Coding Large Language Modelscited by 571
  5. 2017Value-Decomposition Networks For Cooperative Multi-Agent Learningcited by 503
  6. 2020Monotonic Value Function Factorisation for Deep Multi-Agent Reinforcement Learningcited by 485
  7. 2015Deep Recurrent Q-Learning for Partially Observable MDPscited by 465
  8. 2023Text2Reward: Reward Shaping with Language Models for Reinforcement Learningcited by 160
  9. 2024RL-VLM-F: Reinforcement Learning from Vision Language Foundation Model Feedbackcited by 156
  10. 2024Survey on Large Language Model-Enhanced Reinforcement Learning: Concept, Taxonomy, and Methodscited by 116
  11. 2019Multi-Agent Game Abstraction via Graph Attention Neural Networkcited by 82
  12. 2020Learning Structured Communication for Multi-agent Reinforcement Learningcited by 21
2 more
  1. 2023Semantic HELM: A Human-Readable Memory for Reinforcement Learningcited by 17
  2. 2023Guide Your Agent with Adaptive Multimodal Rewardscited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 4.

9 papers.

  1. 2025The Landscape of Agentic Reinforcement Learning for LLMs: A Surveycited by 205
  2. 2025Maestro: Learning to Collaborate via Conditional Listwise Policy Optimization for Multi-Agent LLMscited by 14
  3. 2026Adaptive Collaboration with Humans: Metacognitive Policy Optimization for Multi-Agent LLMs with Continual Learningcited by 9
  4. 2025VideoChat-M1: Collaborative Policy Planning for Video Understanding via Multi-Agent Reinforcement Learningcited by 6
  5. 2026Discovering Reinforcement Learning Interfaces with Large Language Modelscited by 2
  6. 2025Proof2Silicon: Prompt Repair for Verified Code and Hardware Generation via Reinforcement Learningcited by 1
  7. 2025Agent2: An Agent-Generates-Agent Framework for Reinforcement Learning Automationcited by 0
  8. 2026Bridging Reasoning and Action: Hybrid LLM-RL Framework for Efficient Cross-Domain Task-Oriented Dialoguecited by 0
  9. 2026ICMAPE: In-Context Multiagent Pure Explorationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026LangMARL: Natural Language Multi-Agent Reinforcement Learning
  2. 2026LLM-ALSO: LLM-Driven Adaptive Learning-Signal Optimization for Multi-Agent Reinforcement Learning
  3. 2025End-to-End Optimization of LLM-Driven Multi-Agent Search Systems via Heterogeneous-Group-Based Reinforcement Learning
  4. 2025Learn as Individuals, Evolve as a Team: Multi-agent LLMs Adaptation in Embodied Environments
  5. 2025RAGEN: Understanding Self-Evolution in LLM Agents via Multi-Turn Reinforcement Learning
  6. 2025Multi-Agent Evolve: LLM Self-Improve through Co-evolution
  7. 2026Rethinking Agentic Reinforcement Learning In Large Language Models
  8. 2026Multi-Agent LLMs for Adaptive Acquisition in Bayesian Optimization