May 3, 2023 front

ML Newsnew | past | best | rss | submit

Stories from May 3, 2023 (UTC)

Go back a day, month, or year. Go forward a day.

1.	SparseGPT: Language Models Can Be Accurately Pruned in One-Shot (arxiv.org)
	211 points by tosh on May 3, 2023 \| hide \| past \| pdf \| 62 comments
2.	Poisoning Language Models During Instruction Tuning (arxiv.org)
	86 points by hardmaru on May 3, 2023 \| hide \| past \| pdf \| 4 comments
3.	Unlimiformer: Long-Range Transformers with Unlimited Length Input (arxiv.org)
	4 points by ftxbro on May 3, 2023 \| hide \| past \| pdf \| discuss
4.	Unlimiformer: Long-Range Transformers with Unlimited Length Input (arxiv.org)
	4 points by goodmachine on May 3, 2023 \| hide \| past \| pdf \| discuss
5.	Accelerating Neural Self-Improvement via Bootstrapping (arxiv.org)
	3 points by ftxbro on May 3, 2023 \| hide \| past \| pdf \| discuss
6.	Dissecting Recall of Factual Associations in Auto-Regressive Language Models (arxiv.org)
	3 points by tim_sw on May 3, 2023 \| hide \| past \| pdf \| discuss
7.	Can LMs Learn New Entities from Descriptions? (arxiv.org)
	2 points by eliseomartelli on May 3, 2023 \| hide \| past \| pdf \| discuss
8.	WizardLM: Empowering Large Language Models to Follow Complex Instructions (arxiv.org)
	2 points by ftxbro on May 3, 2023 \| hide \| past \| pdf \| discuss