July 22, 2025 front

ML Newsnew | past | best | rss | submit

Stories from July 22, 2025 (UTC)

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

1.	Hierarchical Reasoning Model (arxiv.org)
	9 points by fofoz on Jul 22, 2025 \| hide \| past \| pdf \| discuss
2.	Liking Yellow Imply Driving a School Bus? Semantic Leakage in LLMs (arxiv.org)
	5 points by Bluestein on Jul 22, 2025 \| hide \| past \| pdf \| discuss
3.	Diffusion Beats Autoregressive in Data-Constrained Settings (arxiv.org)
	5 points by georgehe9 on Jul 22, 2025 \| hide \| past \| pdf \| discuss
4.	Gemini 2.5 Pro Capable of Winning Gold at IMO 2025 (arxiv.org)
	4 points by __rito__ on Jul 22, 2025 \| hide \| past \| pdf \| discuss
5.	Diffusion Beats Autoregressive in Data-Constrained Settings (arxiv.org)
	3 points by badmonster on Jul 22, 2025 \| hide \| past \| pdf \| 1 comment
6.	Safety Evaluations of 20 LLMs (arxiv.org)
	1 point by gyanveda on Jul 22, 2025 \| hide \| past \| pdf \| 1 comment
7.	g-AMIE: Towards a Safer Medical AI (arxiv.org)
	1 point by oldfuture on Jul 22, 2025 \| hide \| past \| pdf \| 1 comment
8.	Machine Bullshit: Characterizing the Emergent Disregard for Truth in LLMs (arxiv.org)
	1 point by ilt on Jul 22, 2025 \| hide \| past \| pdf \| discuss