作者位次 #2/5ICML 2026Poster· Deep Learning / Large Language Models
The Geometry of Narrow Fine-Tuning Degradation: Trajectory Lock-in and Spectral Bifurcation
Jia Liu · Jiaxin Luo · Xinhao Qiu · Yixue Hao · Min Chen
强化学习与后训练OpenReview
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。
Jia Liu · Jiaxin Luo · Xinhao Qiu · Yixue Hao · Min Chen
Jia Liu · Jiaxin Luo · Weiwen Xu · Jonathan M. Garibaldi · Xiao-Kun Wu · Yixue Hao · Min Chen
Chain-of-Thought (CoT) prompting enables large language models to produce multi-step reasoning, yet how such reasoning-related structure is expressed during training remains poorly understood. We present Gradient-based Structural Developer (GSD), an unsupervised framework with a principled gradie...
Jia Liu · Jiaxin Luo · Lejun Ai · Yue Wang · Enpeng Lan · Yulong Li · Zongyu Li · Fuxin Zhang · Siyuan Liu · Long Hu · Yixue Hao · Min Chen