一作ICML 2026Poster· Reinforcement Learning / Batch/Offline
Compositional Transduction with Latent Analogies for Offline Goal-Conditioned Reinforcement Learning
Junseok Kim · Dohyeong Kim · Mineui Hong · Songhwai Oh
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Junseok Kim · Dohyeong Kim · Mineui Hong · Songhwai Oh
Junseok Kim · Nakyeong Yang · Kyungmin Min · Kyomin Jung
Self-Consistency improves reasoning reliability through multi-sample aggregation, but incurs substantial inference cost. Adaptive self-consistency methods mitigate this issue by adjusting the sampling budget; however, they rely on count-based stopping rules that treat all responses equally, often...