New Research Advances Multi-Agent Reinforcement Learning Across Cooperation, LLM Agents, and Mobility Policy

Three new papers propose MARL methods: epigraph-form guided exploration for cooperative behavior, a latent-KV architecture for multi-agent LLM systems, and adaptive penalties to regulate autonomous mobility-on-demand operators.

Read the full story at arXiv cs.AI ↗

Timeline · 3 reports

1. 2026-11-06 05:00 · arXiv cs.LG
   Regulating Autonomous Mobility-on-Demand Operators with Fixed and Adaptive Penalties: A Multi-Agent Reinforcement Learning Approach
2. 2026-11-06 05:00 · arXiv cs.LG
   Octopus: One Brain, Many Branches for Latent-KV Multi-Agent Reinforcement Learning
3. 2026-11-06 05:00 · arXiv cs.AI
   Cooperative Multi-Agent Reinforcement Learning via Epigraph-Form Guided Exploration