October 10, 2025 front

ML Newsnew | past | best | rss | submit

Stories from October 10, 2025 (UTC)

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

1.	Reasoning LLMs are wandering solution explorers (arxiv.org)
	90 points by Surreal4434 359 days ago \| hide \| past \| pdf \| 98 comments
2.	Barbarians at the Gate: How AI Is Upending Systems Research (arxiv.org)
	8 points by qianli_cs 359 days ago \| hide \| past \| pdf \| discuss
3.	Less is More: An LLM that outscores Claude Sonnet 4 while being 50.000x smaller (arxiv.org)
	5 points by llosio 358 days ago \| hide \| past \| pdf \| 1 comment
4.	The Missing Link Between the Transformer and Models of the Brain (arxiv.org)
	4 points by Labo333 359 days ago \| hide \| past \| pdf \| 1 comment
5.	Advancing medical artificial intelligence using a century of cases (arxiv.org)
	3 points by hhs 359 days ago \| hide \| past \| pdf \| 1 comment
6.	Moloch's Bargain: Emergent misalignment when LLMs compete for audiences (arxiv.org)
	2 points by felineflock 358 days ago \| hide \| past \| pdf \| discuss
7.	LLMs Reproduce Human Purchase Intent via Semantic Similarity (arxiv.org)
	2 points by anavette 358 days ago \| hide \| past \| pdf \| 1 comment
8.	H1: Bootstrapping LLMs to Reason over Longer Horizons via Reinforcement Learning (arxiv.org)
	2 points by saynotocoffee 359 days ago \| hide \| past \| pdf \| discuss
9.	Truth-Aware Decoding: Program Logic for Factual LMs (arxiv.org)
	2 points by HenryAI 359 days ago \| hide \| past \| pdf \| 1 comment
10.	Less is More: An LLM that outscores Claude Sonnet 4 while being 50.000x smaller (arxiv.org)
	2 points by llosio 359 days ago \| hide \| past \| pdf \| discuss
11.	Bad acronyms in papers are amusing (arxiv.org)
	2 points by AntoineN2 359 days ago \| hide \| past \| pdf \| discuss
12.	New paper: A single character can make or break your LLM evals (arxiv.org)
	1 point by mark_yellow 359 days ago \| hide \| past \| pdf \| 1 comment