Research map / Multi-agent systems and games / Cooperative multi-agent RL
Negotiated Reasoning: On Provably Addressing Relative Over-Generalization
Junjie Sheng, Yantian Wang, Bo Jin, Hongyuan Zha, Jun Wang, Wenhao Li, Xiangfeng Wang
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| condition each agent on previous agents' negotiated actions (nested negotiation, vs none SVNR-M) (ablation) | SVNR · Max of Three (s2 = 1.5, 2.0, 3.0) | training reward | Improved |
| condition each agent on previous agents' negotiated actions (nested negotiation, vs none SVNR-M) (ablation) | SVNR · Particle Gather | training reward | Improved |
| condition on all other agents (full negotiation SVNR-F) instead of strict nested (ablation) | SVNR · Max of Three | training reward | No clear effect |
| use SVNR instead of MASQL/PR2/ROMMEO/L-MADRL | SVNR · Max Of Three (s2=1.5) | test execution return: -0.07 -> 9.71 (vs L-MADRL) | Improved |
| use SVNR instead of L-MADRL | SVNR · Max Of Three (s2=3.0) | test execution return: 9.54 -> 9.60 (L-MADRL vs SVNR) | No clear effect |
| use SVNR instead of baselines | SVNR · Particle Gather | test execution return: 0.00 -> 4.76 (vs MADDPG) | Improved |
| use SVNR to capture both modes | SVNR · Two Modalities | converged perceived joint policy | Improved |
Closest papers by meaning
- Optimistic Multi-Agent Policy Gradient (ICML 2024 · similarity 0.88)
- Cooperative Policy Agreement: Learning Diverse Policy for Offline MARL (AAAI 2025 · similarity 0.87)
- Dr. MAS: Stable Reinforcement Learning for Multi-Agent LLM Systems (NeurIPS 2026 · similarity 0.87)
- Adaptive Collaboration with Humans: Metacognitive Policy Optimization for Multi-Agent LLMs with Continual Learning (ICLR 2026 · similarity 0.86)
- Stronger-MAS: Multi-Agent Reinforcement Learning for Collaborative LLMs (ICLR 2026 · similarity 0.86)
- From Debate to Equilibrium: Belief‑Driven Multi‑Agent LLM Reasoning via Bayesian Nash Equilibrium (ICML 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
