January 22, 2025 front

ML Newsnew | past | best | rss | submit

Stories from January 22, 2025 (UTC)

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

1.	Tensor Product Attention Is All You Need (arxiv.org)
	160 points by eunos on Jan 22, 2025 \| hide \| past \| pdf \| 104 comments
2.	Lossless Compression of Vector IDs for Approximate Nearest Neighbor Search (arxiv.org)
	151 points by fzliu on Jan 22, 2025 \| hide \| past \| pdf \| 6 comments
3.	Flame: A small language model for spreadsheet formulas (2023) (arxiv.org)
	117 points by azhenley on Jan 22, 2025 \| hide \| past \| pdf \| 18 comments
4.	Can LLMs demonstrate behavioral self-awareness? (arxiv.org)
	3 points by omarsar on Jan 22, 2025 \| hide \| past \| pdf \| 1 comment
5.	VideoWorld: Exploring Knowledge Learning from Unlabeled Video (arxiv.org)
	2 points by TaurenHunter on Jan 22, 2025 \| hide \| past \| pdf \| discuss
6.	Test-time regression: a unifying framework for designing sequence models (arxiv.org)
	1 point by ketothekingdom on Jan 22, 2025 \| hide \| past \| pdf \| discuss
7.	Evolving Deeper LLM Thinking (arxiv.org)
	1 point by simonpure on Jan 22, 2025 \| hide \| past \| pdf \| discuss
8.	Physics of Skill Learning (arxiv.org)
	1 point by elashri on Jan 22, 2025 \| hide \| past \| pdf \| discuss