arXiv Machine Learning By Itai Morad, Nir Shlezinger, Yonina C. Eldar

SGD-Based Knowledge Distillation with Bayesian Teachers: Theory and Guidelines

Read the original on arXiv Machine Learning →

arXiv:2601. 01484v2 Announce Type: replace Abstract: Knowledge Distillation (KD) is a central paradigm for transferring knowledge from a large teacher network to a typically smaller student model, often by leveraging soft probabilistic outputs.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv AI
Jul 29

Rethinking Likelihood distributions: Student's t Likelihood Boosts Bayesian Neural Network Performance

arXiv:2607. 25376v1 Announce Type: cross Abstract: In Bayesian neural networks (BNNs), variational inference is a widely adopted framework for modeling uncertainty in a distributional way, with the evidence lower bound (ELBO) serving as the standard objective function.

By Pei-Hsuan Hsia, Lars H. Heyen, Arvid Weyrauch, Markus Goetz, Achim Streit, Sebastian Krumscheid, Charlotte Debus
arXiv AI
Jul 3

DemoPSD: Disagreement-Modulated Policy Self-Distillation

arXiv:2607. 02502v1 Announce Type: cross Abstract: On-policy self-distillation (OPSD) has emerged as a practical method for training large language models (LLMs) to reason, where a single model acts as both the teacher and the student with different levels of information access.

By Yunhe Li, Hao Shi, Wenhao Liu, Mengzhe Ruan, Hanxu Hou, Zhongxiang Dai, Shuang Qiu, Linqi Song