AI RESEARCH
Benchmarking the Limits of In-Context Reinforcement Learning for Ad-Hoc Teamwork
arXiv CS.AI
•
ArXi:2605.24423v1 Announce Type: new In-Context Reinforcement Learning (ICRL) has enabled foundation agents to adapt instantaneously to novel tasks, yet its efficacy in Ad-Hoc Teamwork (AHT)-where coordination with unknown partners is required-remains unexplored. To rigorously evaluate this, we