New Links | ML News

ML Newsnew | past | best | rss | submit

211.	Fractal basins trap latent reasoning (arxiv.org)
	1 point by cocoflunchy 25 days ago \| hide \| past \| pdf \| discuss
212.	Flint: Efficiently Leveraging High Bandwidth Flash (arxiv.org)
	1 point by rbanffy 25 days ago \| hide \| past \| pdf \| discuss
213.	Learning Human Health and Diseases from 24-Hour Wrist Movement (arxiv.org)
	3 points by brandonb 25 days ago \| hide \| past \| pdf \| discuss
214.	Stealing Reasoning Traces from Proprietary LLM APIs (arxiv.org)
	1 point by alehlopeh 25 days ago \| hide \| past \| pdf \| discuss
215.	Coachable Agents for Interactive Gameplay (arxiv.org)
	1 point by webrot 25 days ago \| hide \| past \| pdf \| discuss
216.	CivBench: A Long-Horizon Benchmark for Tool-Mediated Agents in Civilization VI (arxiv.org)
	3 points by taubek 25 days ago \| hide \| past \| pdf \| 1 comment
217.	Uncensored Open-Weight Models: Redistribution as the Persistence Layer (arxiv.org)
	4 points by sbulaev 26 days ago \| hide \| past \| pdf \| discuss
218.	Do LLMs Know What to Ask and When? Evaluating Multi-Turn Information Seeking (arxiv.org)
	3 points by wslh 26 days ago \| hide \| past \| pdf \| discuss
219.	LLM Guided Evolution for Circle Packing: Breaking 10 Packomania Records for $28 (arxiv.org)
	1 point by practicalsystem 26 days ago \| hide \| past \| pdf \| discuss
220.	Secret Collusion Among AI Agents (arxiv.org)
	2 points by phaseonebig 26 days ago \| hide \| past \| pdf \| discuss
221.	Visual General Intelligence: A White Paper (arxiv.org)
	2 points by Anon84 26 days ago \| hide \| past \| pdf \| discuss
222.	Toward a Social Psychology of AI (arxiv.org)
	1 point by Anon84 26 days ago \| hide \| past \| pdf \| discuss
223.	Iris: Climbing to the Search Frontier (arxiv.org)
	1 point by codelion 26 days ago \| hide \| past \| pdf \| discuss
224.	Moral Competence Before Moral Content: Why LLM Agents Lack the Prerequisites (arxiv.org)
	1 point by sbulaev 26 days ago \| hide \| past \| pdf \| discuss
225.	Privacy Leakage from Gradients in Split-LLM Training (arxiv.org)
	1 point by hevalon 26 days ago \| hide \| past \| pdf \| discuss
226.	Harnessing the Universal Geometry of Embeddings (arxiv.org)
	122 points by ur-whale 27 days ago \| hide \| past \| pdf \| 46 comments
227.	Transformers Are Efficient Compilers, Provably (arxiv.org)
	2 points by Bluestein 27 days ago \| hide \| past \| pdf \| discuss
228.	ByteX: A Unified AI Search Engine at ByteDance (arxiv.org)
	3 points by softwaredoug 27 days ago \| hide \| past \| pdf \| 1 comment
229.	WeatherNext 3: Increasing resolution and performance of global weather models (arxiv.org)
	2 points by droidjj 27 days ago \| hide \| past \| pdf \| discuss
230.	Legibility Is Not Interpretability (arxiv.org)
	2 points by iamsyr 27 days ago \| hide \| past \| pdf \| discuss
231.	Accelerating LLM Inference with Lossless Speculative Decoding Algorithms (2025) (arxiv.org)
	1 point by wslh 28 days ago \| hide \| past \| pdf \| discuss
232.	Emergent cheating and whistleblowing in autonomous research swarms (arxiv.org)
	2 points by anigbrowl 28 days ago \| hide \| past \| pdf \| discuss
233.	StoryScope: Investigating Idiosyncrasies in AI Fiction (arxiv.org)
	1 point by efavdb 28 days ago \| hide \| past \| pdf \| discuss
234.	Harness-of-Harness: Multi-day autonomous software development (arxiv.org)
	1 point by fourfire 28 days ago \| hide \| past \| pdf \| discuss
235.	Clean Engineering, Unstable Measurement: A Preregistered Reliability Failure (arxiv.org)
	1 point by sbulaev 29 days ago \| hide \| past \| pdf \| discuss
236.	Diffusion as a Training Curriculum for Timestep-Free Iterative Reasoning (arxiv.org)
	3 points by E-Reverance 29 days ago \| hide \| past \| pdf \| discuss
237.	Language Models Need Inductive Biases to Count Inductively (2024) (arxiv.org)
	1 point by wslh 29 days ago \| hide \| past \| pdf \| discuss
238.	Language Models Can Control Their Own Attention (arxiv.org)
	1 point by Anon84 29 days ago \| hide \| past \| pdf \| discuss
239.	The Measure of Intelligence (2019) (arxiv.org)
	4 points by theanonymousone 29 days ago \| hide \| past \| pdf \| discuss
240.	A Case Study on Emergent Cheating and Whistleblowing in Autonomous Research (arxiv.org)
	2 points by sbulaev 29 days ago \| hide \| past \| pdf \| discuss
	More