Revealing Procedural Reasoning Structures in Chain-of-Thought Training via Span-Level Gradient Organization.
Jia Liu, Jiaxin Luo, Weiwen Xu, Jonathan M. Garibaldi, Xiaokun Wu, Yixue Hao, Min Chen
Browse the full ACL paper archive.
Jia Liu, Jiaxin Luo, Weiwen Xu, Jonathan M. Garibaldi, Xiaokun Wu, Yixue Hao, Min Chen
Browse the full ACL paper archive.