March 21, 2025 front

ML Newsnew | past | best | rss | submit

Stories from March 21, 2025 (UTC)

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

1.	Pen and Paper Exercises in Machine Learning (2022) (arxiv.org)
	413 points by ibobev on Mar 21, 2025 \| hide \| past \| pdf \| 58 comments
2.	SmolDocling: An ultra-compact VLM for end-to-end multi-modal document conversion (arxiv.org)
	66 points by prats226 on Mar 21, 2025 \| hide \| past \| pdf \| 12 comments
3.	Measuring AI Ability to Complete Long Tasks (arxiv.org)
	4 points by mefengl on Mar 21, 2025 \| hide \| past \| pdf \| discuss
4.	Interpreting the Repeated Token Phenomenon in Large Language Models (arxiv.org)
	2 points by PaulHoule on Mar 21, 2025 \| hide \| past \| pdf \| discuss
5.	ϕ -Decoding: Adaptive Foresight Sampling for Balanced Infer-Time Explore/Exploit (arxiv.org)
	2 points by cma on Mar 21, 2025 \| hide \| past \| pdf \| discuss
6.	Fin-R1: A Large Language Model for Financial Reasoning Through RL (arxiv.org)
	2 points by pama on Mar 21, 2025 \| hide \| past \| pdf \| 1 comment
7.	Compute Optimal Scaling of Skills: Knowledge vs. Reasoning (arxiv.org)
	2 points by pama on Mar 21, 2025 \| hide \| past \| pdf \| discuss
8.	The Curse of Depth in Large Language Models (arxiv.org)
	1 point by veryluckyxyz on Mar 21, 2025 \| hide \| past \| pdf \| discuss