Hugging Face Trending Papers

LP-SFT: Local-Preserving Supervised Fine-Tuning via Multimodal Entropy Structure

Read the original on Hugging Face Trending Papers →

Supervised fine-tuning (SFT) is the standard approach for adapting pretrained language models to downstream domains, yet it often improves target-domain behavior at the cost of degrading pre-existing capabilities. Standard cross-entropy fine-tuning promotes only the observed label token and leaves unconstrained how probability mass is redistributed over other plausible alternatives, potentially distorting the rich local preference structure learned during pretraining.

Summary generated by The Flow from the publisher's feed. The full article lives at Hugging Face Trending Papers.

arXiv Machine Learning
6d ago

Weightless Fine-Tuning: Personalizing LLMs via Logit-Space Transport

arXiv:2608. 11342v1 Announce Type: new Abstract: Supervised fine-tuning (SFT) is a standard approach for adapting LLMs to a target distribution, but in settings such as personalization, where each author requires separate weight access, optimization, storage, and retraining, its costs become prohibitive.

By Bohan Zhang, Anqi Ni, Yixin Wang, Paramveer S. Dhillon