March 7, 2025 front

ML Newsnew | past | best | rss | submit

Stories from March 7, 2025 (UTC)

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

1.	Ladder: Self-improving LLMs through recursive problem decomposition (arxiv.org)
	370 points by fofoz on Mar 7, 2025 \| hide \| past \| pdf \| 110 comments
2.	Hot: Highlighted Chain of Thought for Referencing Supporting Facts from Inputs (arxiv.org)
	4 points by taesiri on Mar 7, 2025 \| hide \| past \| pdf \| 1 comment
3.	RingFormer: Rethinking Recurrent Transformer with Adaptive Level Signals (arxiv.org)
	3 points by PaulHoule on Mar 7, 2025 \| hide \| past \| pdf \| discuss
4.	Computation-Aware ControlNet with Dynamic Router for Text-to-Image Generation (arxiv.org)
	3 points by jinqueeny on Mar 7, 2025 \| hide \| past \| pdf \| discuss
5.	L1: Controlling How Long a Reasoning Model Thinks with Reinforcement Learning (arxiv.org)
	2 points by papers2092 on Mar 7, 2025 \| hide \| past \| pdf \| discuss
6.	Think Inside the JSON: Reinforcement Strategy for Strict LLM Schema Adherence (arxiv.org)
	1 point by PaulHoule on Mar 7, 2025 \| hide \| past \| pdf \| discuss
7.	Deep Learning Is Not So Mysterious or Different (arxiv.org)
	1 point by thoughtpeddler on Mar 7, 2025 \| hide \| past \| pdf \| discuss