CommentsInternational Conference on Machine Learning (ICML), 2026. v4: Revised CPC theory to (a) show enforced smoothness for likelihood-ratio control parameter and (b) for general control parameter, assume smoothness only of constrained policy $π_t^{(β)}$ w.r.t. $β$ (rather than of $(l_i - α)\cdotπ_t^{(β)}$ or of conformal weights)
Comments171 pages. Formalized in Lean 4 with Mathlib: 240 theorems in the elaborated environment, 141 audited headline results, cold-compiling from a clean checkout with zero custom axioms. Source, theorem-by-theorem contract, and reproducible axiom audit: https://github.com/selfreferencing/TSE_Formal. Companion to Agentic Capital
机构
*
Clemson University(克莱姆森大学)
;
LinkedIn(领英)
;
Washington University in St. Louis(圣路易斯华盛顿大学)
;
Arizona State University(亚利桑那州立大学)
;
Columbia University(哥伦比亚大学)
CommentsThis preprint has not undergone peer review or any post-submission improvements or corrections. The Version of Record of this contribution is published in Artificial Intelligence in HCI (HCII 2026), Lecture Notes in Computer Science, vol. 16745, and is available online at https://doi.org/10.1007/978-3-032-30849-8_13
Journal refDegen, H., Ntoa, S. (eds) Artificial Intelligence in HCI. HCII 2026. Lecture Notes in Computer Science, vol 16745, pp. 219--242. Springer, Cham (2026)
机构
*
Transitional Artificial Intelligence Research Group, School of Mathematics and Statistics(过渡人工智能研究组,数学与统计学学院)
;
Centre for Artificial Intelligence and Innovation(人工智能与创新中心)
;
Pingla Institute(Pingla研究所)
Entropy-Aware On-Policy Distillation of Language Models
熵感知的在线策略蒸馏语言模型
Woogyeol Jin, Taywon Min, Yongjin Yang, Dennis Wei, Yi Zhou, Swanand Ravindra Kadhe, Nathalie Baracaldo, Kimin Lee
机构
*
IBM Research, San Jose, CA, USA(IBM研究院,旧金山,加州,美国)
;
University of Toronto, Ontario, Canada(多伦多大学,安大略,加拿大)
;
Vector Institute, Ontario, Canada(向量研究所,安大略,加拿大)