IEEE

Structured Gradient Guidance for Few-Shot Adaptation in Large Language Models

In recent years, Large Language Models (LLMs) have achieved remarkable success in natural language processing tasks, driving intelligent upgrades in many downstream applications. With strong ...
IEEE

SSFT: Algorithm and Hardware Co-design for Structured Sparse Fine-Tuning of Large Language Models

Abstract: A significant number of users depend on Large Language Models (LLMs) for downstream tasks, but training LLMs from scratch remains prohibitively expensive. Sparse finetuning (SFT) has emerged ...