2 papers
cs.AI2026
Evaluating Risks in Weak-to-Strong Alignment: A Bias-Variance Perspective
Hamid Osooli, Kareema Batool, Rick Gentry +3
Weak-to-strong alignment offers a promising route to scalable supervision, but it can fail when a strong model becomes confidently wrong on examples that lie in the weak model's bl…
cs.MA2026
Zero-Shot Coordination in Ad Hoc Teams with Generalized Policy Improvement and Difference Rewards
Rupal Nigam, Niket Parikh, Hamid Osooli +3
Real-world multi-agent systems may require ad hoc teaming, where an agent must coordinate with other previously unseen teammates to solve a task in a zero-shot manner. Prior work o…