March 30, 2026 front

ML Newsnew | past | best | rss | submit

Stories from March 30, 2026 (UTC)

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

1.	Stanford study reveals AI vision models invent images they never see (arxiv.org)
	49 points by LionTurtle13 188 days ago \| hide \| past \| pdf \| 1 comment
2.	Galton's Law of Mediocrity: Why Large Language Models Regress to the Mean (arxiv.org)
	4 points by camillomiller 188 days ago \| hide \| past \| pdf \| discuss
3.	An Alternative Trajectory for Generative AI (arxiv.org)
	3 points by haltingproblem 187 days ago \| hide \| past \| pdf \| discuss
4.	Did You Forget What I Asked? Memory Failures in Large Language Models (arxiv.org)
	3 points by PaulHoule 187 days ago \| hide \| past \| pdf \| discuss
5.	ARC-AGI-3: A New Challenge for Frontier Agentic Intelligence (arxiv.org)
	2 points by doener 187 days ago \| hide \| past \| pdf \| discuss
6.	An Autonomous Agentic Secure Code Review for Immature Vulnerabilities Detection (arxiv.org)
	2 points by wslh 187 days ago \| hide \| past \| pdf \| discuss
7.	Natural-Language Agent Harnesses (arxiv.org)
	2 points by Jimmc414 187 days ago \| hide \| past \| pdf \| discuss
8.	d-HNSW: A High-Performance Vector Search Engine on Disaggregated Memory (arxiv.org)
	1 point by tanelpoder 187 days ago \| hide \| past \| pdf \| discuss
9.	PixelSmile: Toward Fine-Grained Facial Expression Editing (arxiv.org)
	1 point by ipotapov 188 days ago \| hide \| past \| pdf \| discuss