2 papers
cs.AI2026
Input-Conditioned Search for Efficient Test-Time Scaling of Recursive Reasoners
Haochen Shi, Tianwen Fu, Qianggang Ding +7
People often tackle difficult problems by thinking more deeply or exploring more possibilities. Similarly, recursive reasoners, which reason by repeatedly applying a learned update…
cs.LG2026
From Policy Evaluation to Off-Policy Control: Provable Emergence of In-Context Q-Learning
Yingjun Mo, Yajie Zhao, Fang Kong
Transformers can implement reinforcement-learning algorithms in context, yet how off-policy control emerges within Transformers remains poorly understood. We study Q-learning, a fo…