February 4, 2025 front

ML Newsnew | past | best | rss | submit

Stories from February 4, 2025 (UTC)

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

1.	DeepRAG: Thinking to retrieval step by step for large language models (arxiv.org)
	191 points by fofoz on Feb 4, 2025 \| hide \| past \| pdf \| 29 comments
2.	Querying Databases with Function Calling (arxiv.org)
	3 points by tosh on Feb 4, 2025 \| hide \| past \| pdf \| discuss
3.	Over-Tokenized Transformer: Vocabulary Is Generally Worth Scaling (arxiv.org)
	2 points by famouswaffles on Feb 4, 2025 \| hide \| past \| pdf \| discuss
4.	Self-Improving Transformers Overcome Easy-to-Hard and Length Generalization (arxiv.org)
	2 points by anothermathbozo on Feb 4, 2025 \| hide \| past \| pdf \| discuss
5.	Scalable-Softmax Is Superior for Attention (arxiv.org)
	2 points by jw1224 on Feb 4, 2025 \| hide \| past \| pdf \| 2 comments
6.	Reinforcing Thinking Through Reasoning-Enhanced Reward Models (arxiv.org)
	2 points by PaulHoule on Feb 4, 2025 \| hide \| past \| pdf \| discuss
7.	OmniHuman-1: Scaling-Up of One-Stage Conditioned Human Animation Models (arxiv.org)
	1 point by neom on Feb 4, 2025 \| hide \| past \| pdf \| discuss
8.	Language Models Use Trigonometry to Do Addition (arxiv.org)
	1 point by fofoz on Feb 4, 2025 \| hide \| past \| pdf \| 3 comments