New Links | ML News

ML Newsnew | past | best | rss | submit

2071.	Deep sequence models tend to memorize geometrically; it is unclear why (arxiv.org)
	4 points by amichail 333 days ago \| hide \| past \| pdf \| discuss
2072.	Accumulating Context Changes the Beliefs of Language Models (arxiv.org)
	2 points by Anon84 333 days ago \| hide \| past \| pdf \| discuss
2073.	Continuous Autoregressive Language Models (arxiv.org)
	115 points by Anon84 334 days ago \| hide \| past \| pdf \| 10 comments
2074.	Evaluating Probabilistic Reasoning in LLMs Through Language-Only Decision Tasks (arxiv.org)
	1 point by PaulHoule 334 days ago \| hide \| past \| pdf \| 1 comment
2075.	Best-Of-∞: Asymptotic Performance of Test-Time Compute (Infinite Compute Budget) (arxiv.org)
	2 points by SweetSoftPillow 334 days ago \| hide \| past \| pdf \| discuss
2076.	Pre-training under infinite compute (arxiv.org)
	20 points by SweetSoftPillow 334 days ago \| hide \| past \| pdf \| discuss
2077.	Kosmos: An AI Scientist for Autonomous Discovery (arxiv.org)
	60 points by belter 334 days ago \| hide \| past \| pdf \| 20 comments
2078.	AutoCode: LLMs as Problem Setters for Competitive Programming (arxiv.org)
	1 point by PaulHoule 334 days ago \| hide \| past \| pdf \| discuss
2079.	Benchmarking multilingual long-context language models (arxiv.org)
	3 points by sysoleg 334 days ago \| hide \| past \| pdf \| discuss
2080.	Can LLMs Subtract Numbers? (arxiv.org)
	2 points by belter 334 days ago \| hide \| past \| pdf \| discuss
2081.	Towards General Auditory Intelligence (arxiv.org)
	2 points by selimonder 335 days ago \| hide \| past \| pdf \| 1 comment
2082.	Verbalized Sampling: How to Mitigate Mode Collapse and Unlock LLM Diversity (arxiv.org)
	1 point by JnBrymn 335 days ago \| hide \| past \| pdf \| discuss
2083.	Continuous Autoregressive Language Models (arxiv.org)
	3 points by guybedo 335 days ago \| hide \| past \| pdf \| 1 comment
2084.	Cache-to-Cache: Direct Semantic Communication Between Large Language Models (arxiv.org)
	14 points by jonbaer 335 days ago \| hide \| past \| pdf \| discuss
2085.	LLM's Report Subjective Experience Under Self-Referential Processing (arxiv.org)
	1 point by gradus_ad 335 days ago \| hide \| past \| pdf \| discuss
2086.	LLMZip: Lossless Text Compression Using Large Language Models (arxiv.org)
	2 points by jfantl 336 days ago \| hide \| past \| pdf \| 4 comments
2087.	Deep sequence models tend to memorize geometrically (arxiv.org)
	3 points by jonbaer 336 days ago \| hide \| past \| pdf \| discuss
2088.	Streaming DiLoCo: Towards a Distributed Free Lunch (arxiv.org)
	2 points by nathan-barry 336 days ago \| hide \| past \| pdf \| discuss
2089.	Kimi Linear: An Expressive, Efficient Attention Architecture (arxiv.org)
	6 points by birriel 337 days ago \| hide \| past \| pdf \| discuss
2090.	Defeating the Training-Inference Mismatch via FP16 (arxiv.org)
	1 point by billyzs 337 days ago \| hide \| past \| pdf \| discuss
2091.	R2T: Rule-Encoded Loss Functions for Low-Resource Sequence Tagging (arxiv.org)
	4 points by PaulHoule 337 days ago \| hide \| past \| pdf \| discuss
2092.	Humains-Junior: A 3.8B Language Model Achieving GPT-4o-Level Factual Accuracy (arxiv.org)
	2 points by gidellav 338 days ago \| hide \| past \| pdf \| discuss
2093.	Watermarking for Generative AI (arxiv.org)
	17 points by gidellav 338 days ago \| hide \| past \| pdf \| discuss
2094.	Agentic AI Home Energy Management System: Residential Load Scheduling (arxiv.org)
	2 points by simonpure 338 days ago \| hide \| past \| pdf \| 2 comments
2095.	TempoPFN: Synthetic Pre-Training of Linear RNNs for Zero-Shot Timeseries Forecas (arxiv.org)
	1 point by jul8234 338 days ago \| hide \| past \| pdf \| discuss
2096.	Remote Labor Index: Measuring AI Automation of Remote Work (arxiv.org)
	2 points by belter 338 days ago \| hide \| past \| pdf \| discuss
2097.	LLMs Report Subjective Experience Under Self-Referential Processing (arxiv.org)
	3 points by j_crick 339 days ago \| hide \| past \| pdf \| 1 comment
2098.	Defeating the Training-Inference Mismatch via FP16 (arxiv.org)
	2 points by matt_d 339 days ago \| hide \| past \| pdf \| discuss
2099.	Benchmarking On-Device Machine Learning on Apple Silicon with MLX (arxiv.org)
	2 points by PaulHoule 339 days ago \| hide \| past \| pdf \| discuss
2100.	Chain-of-Thought Hijacking (arxiv.org)
	2 points by belter 339 days ago \| hide \| past \| pdf \| discuss
	More