sparse-updates

Tag

Cards List
#sparse-updates

Fine-Tuning of Transformer models with Frames

arXiv cs.AI · 2026-08-28 Cached

The paper proposes FrameFT, a parameter-efficient fine-tuning method using sparse coefficients in a Fusion Frame basis, reducing memory footprint while achieving performance on par with or exceeding state-of-the-art PEFT techniques.

0 favorites 0 likes
#sparse-updates

LongAct: Harnessing Intrinsic Activation Patterns for Long-Context Reinforcement Learning

Hugging Face Daily Papers · 2026-04-16 Cached

LongAct proposes a saliency-guided sparse update strategy for improving long-context reasoning in LLMs by selectively updating weights associated with high-magnitude activations in query and key vectors, achieving ~8% improvement on LongBench v2.

0 favorites 0 likes
← Back to home

Submit Feedback