{
  "id": 635,
  "url": "https://arxiv.org/abs/2606.25073v1",
  "title": "GCT-MARL: Graph-Based Contrastive Transfer for Sample-Efficient Cooperative Multi-Agent Reinforcement Learning",
  "summary": "In cooperative multi-agent reinforcement learning (MARL), from a deployment perspective, it is challenging and expensive to train agents from scratch for each new environment or task. In this work, we propose GCT-MARL, a transfer learning framework that builds on the multi-view graph contrastive backbone of MAIL and augments it with a per-view, adaptively weighted alignment loss and a two-phase training protocol specifically designed for transfer across populations of varying sizes and compositi",
  "authors": "Animesh Animesh, Satheesh K Perepu, Kaushik Dey",
  "category": "research",
  "topics": "safety-alignment,agents-autonomy,environment",
  "orgs": null,
  "regions": null,
  "published_at": "2026-06-23T18:31:27.000Z",
  "fetched_at": "2026-07-14T14:14:41.550Z",
  "source_slug": "arxiv-ethics",
  "source_name": "arXiv",
  "source_homepage": "https://arxiv.org",
  "ethics_ai_record_url": "https://ethics.ai/record/635",
  "original_url": "https://arxiv.org/abs/2606.25073v1",
  "evidence_status": "source-only",
  "attribution": "via ethics.ai"
}