July 23, 2025 front

ML Newsnew | past | best | rss | submit

Stories from July 23, 2025 (UTC)

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

1.	CT-ScanGaze: A Dataset and Baselines for 3D Volumetric Scanpath Modeling (arxiv.org)
	4 points by PaulHoule on Jul 23, 2025 \| hide \| past \| pdf \| discuss
2.	Prompt Injection 2.0: Hybrid AI Threats – Paper and Open Source Testing Toolkit (arxiv.org)
	3 points by jmchugh9 on Jul 23, 2025 \| hide \| past \| pdf \| 1 comment
3.	BrainScaleS: A Wafer-Scale, Neuromorphic System [pdf] (arxiv.org)
	3 points by nickpsecurity on Jul 23, 2025 \| hide \| past \| pdf \| 1 comment
4.	Deep Researcher with Test-Time Diffusion (arxiv.org)
	3 points by omarsar on Jul 23, 2025 \| hide \| past \| pdf \| discuss
5.	UnMarker: A Universal Attack on Defensive Image Watermarking (arxiv.org)
	2 points by gnabgib on Jul 23, 2025 \| hide \| past \| pdf \| discuss
6.	Weak baselines and reporting biases lead to overoptimism in ML for fluid PDEs (arxiv.org)
	2 points by MITSardine on Jul 23, 2025 \| hide \| past \| pdf \| discuss
7.	Improving KAN with CDF normalization to quantiles (arxiv.org)
	1 point by jarekd on Jul 23, 2025 \| hide \| past \| pdf \| 2 comments