May 6, 2025 front

ML Newsnew | past | best | rss | submit

Stories from May 6, 2025 (UTC)

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

1.	DoomArena: A Framework for Testing AI Agents Against Evolving Security Threats (arxiv.org)
	10 points by PaulHoule on May 6, 2025 \| hide \| past \| pdf \| 2 comments
2.	Don't be lazy: CompleteP enables compute-efficient deep transformers (arxiv.org)
	4 points by nsdey on May 6, 2025 \| hide \| past \| pdf \| discuss
3.	Towards Dataset Copyright Evasion Attack Against Personalized Diffusion Models (arxiv.org)
	3 points by badmonster on May 6, 2025 \| hide \| past \| pdf \| discuss
4.	Unveiling the Hidden: Movie Genre and User Bias in Spoiler Detection (arxiv.org)
	3 points by PaulHoule on May 6, 2025 \| hide \| past \| pdf \| discuss
5.	(How) Do reasoning models reason? (arxiv.org)
	3 points by YeGoblynQueenne on May 6, 2025 \| hide \| past \| pdf \| discuss
6.	Language Representations Can Be What Recommenders Need: Findings and Potentials (arxiv.org)
	2 points by PaulHoule on May 6, 2025 \| hide \| past \| pdf \| discuss
7.	HalluMix Benchmark: Detecting Hallucinations in Real-World Scenarios (arxiv.org)
	2 points by rancar2 on May 6, 2025 \| hide \| past \| pdf \| discuss
8.	Evaluating Frontier Models for Stealth and Situational Awareness (arxiv.org)
	2 points by badmonster on May 6, 2025 \| hide \| past \| pdf \| discuss
9.	Cost-of-Pass: An Economic Framework for Evaluating Language Models (arxiv.org)
	1 point by JumpCrisscross on May 6, 2025 \| hide \| past \| pdf \| discuss