June 17, 2024 front

ML Newsnew | past | best | rss | submit

Stories from June 17, 2024 (UTC)

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

1.	Creativity has left the chat: The price of debiasing language models (arxiv.org)
	174 points by hardmaru on Jun 17, 2024 \| hide \| past \| pdf \| 225 comments
2.	An Image Is Worth 32 Tokens for Reconstruction and Generation [pdf] (arxiv.org)
	3 points by croes on Jun 17, 2024 \| hide \| past \| pdf \| discuss
3.	Large Language Models for Forecasting and Anomaly Detection (arxiv.org)
	2 points by rntn on Jun 17, 2024 \| hide \| past \| pdf \| discuss
4.	Neural Thermodynamic Integration: Free Energies from Diffusion Models (arxiv.org)
	2 points by jasondavies on Jun 17, 2024 \| hide \| past \| pdf \| discuss
5.	AlphaMath Almost Zero: process Supervision without process (arxiv.org)
	2 points by hardmaru on Jun 17, 2024 \| hide \| past \| pdf \| discuss
6.	The Factorization Curse: Which Tokens You Predict Underlie the Reversal Curse (arxiv.org)
	1 point by PaulHoule on Jun 17, 2024 \| hide \| past \| pdf \| discuss
7.	Render and Diffuse: Aligning Image and Action for Behaviour Cloning (arxiv.org)
	1 point by geox on Jun 17, 2024 \| hide \| past \| pdf \| discuss
8.	Large Language Model Confidence Estimation via Black-Box Access (arxiv.org)
	1 point by PaulHoule on Jun 17, 2024 \| hide \| past \| pdf \| discuss