New Links | ML News

ML Newsnew | past | best | rss | submit

1111.	Estimating Black-Box LLM Parameter Counts via Factual Capacity (arxiv.org)
	3 points by RockstarSprain 157 days ago \| hide \| past \| pdf \| discuss
1112.	Beyond 80/20: High-Entropy Minority Tokens Drive Effective RL for LLM Reasoning (arxiv.org)
	3 points by mdp2021 157 days ago \| hide \| past \| pdf \| discuss
1113.	Training Large Language Models to Reason in a Continuous Latent Space [pdf] (arxiv.org)
	1 point by thunderbong 158 days ago \| hide \| past \| pdf \| discuss
1114.	OpenGame: Open Agentic Coding for Games (arxiv.org)
	2 points by lexandstuff 159 days ago \| hide \| past \| pdf \| discuss
1115.	From Skills to Talent: Organising Heterogeneous Agents as a Company [pdf] (arxiv.org)
	2 points by SerCe 159 days ago \| hide \| past \| pdf \| discuss
1116.	Evaluating CUDA Tile for AI Workloads on Hopper and Blackwell GPUs (arxiv.org)
	2 points by matt_d 159 days ago \| hide \| past \| pdf \| discuss
1117.	Learning to Orchestrate Agents in Natural Language with the Conductor (arxiv.org)
	1 point by Anon84 159 days ago \| hide \| past \| pdf \| discuss
1118.	Universal Transformers Need Memory: Depth-State Trade-Offs in Adaptive Recursive (arxiv.org)
	1 point by che_shr_cat 159 days ago \| hide \| past \| pdf \| discuss
1119.	The Limits of Self-Improving in Large Language Models (arxiv.org)
	1 point by darccio 159 days ago \| hide \| past \| pdf \| discuss
1120.	Image Generators Are Generalist Vision Learners (arxiv.org)
	2 points by gmays 159 days ago \| hide \| past \| pdf \| discuss
1121.	Does Point Cloud Boost Spatial Reasoning of Large Language Models? (arxiv.org)
	1 point by gregsadetsky 159 days ago \| hide \| past \| pdf \| discuss
1122.	Scaling Test-Time Compute for Agentic Coding (arxiv.org)
	2 points by gmays 159 days ago \| hide \| past \| pdf \| discuss
1123.	LLMs Corrupt Your Documents When You Delegate (arxiv.org)
	2 points by interpol_p 159 days ago \| hide \| past \| pdf \| discuss
1124.	AI prefers resumes written by itself: Self-preferencing in Algorithmic Hiring (arxiv.org)
	3 points by ytpete 159 days ago \| hide \| past \| pdf \| 1 comment
1125.	Generation Is Required for Data-Efficient Perception (arxiv.org)
	1 point by E-Reverance 160 days ago \| hide \| past \| pdf \| discuss
1126.	Microsoft TRELLIS.2: An Open-Source, 4B-Parameter, Image-to-3D Model [pdf] (arxiv.org)
	1 point by thunderbong 160 days ago \| hide \| past \| pdf \| discuss
1127.	ASI-Evolve: AI Accelerates AI (arxiv.org)
	1 point by champagnepapi 160 days ago \| hide \| past \| pdf \| discuss
1128.	Microsoft Paper: LLMs Corrupt Your Documents When You Delegate (Arxiv.org) (arxiv.org)
	7 points by wuschel 160 days ago \| hide \| past \| pdf \| 2 comments
1129.	Guess-Verify-Refine: Data-Aware Top-K for Sparse-Attention Decoding on Blackwell (arxiv.org)
	4 points by matt_d 160 days ago \| hide \| past \| pdf \| 1 comment
1130.	The Platonic Representation Hypothesis (arxiv.org)
	2 points by Anon84 160 days ago \| hide \| past \| pdf \| 1 comment
1131.	Learning to Repair Lean Proofs from Compiler Feedback (arxiv.org)
	1 point by matt_d 161 days ago \| hide \| past \| pdf \| discuss
1132.	The Quantization Robustness of Diffusion Language Models in Coding Benchmarks (arxiv.org)
	3 points by matt_d 161 days ago \| hide \| past \| pdf \| discuss
1133.	Beyond Silicon: Materials, Mechanisms, and Methods for Physical Neural Computing (arxiv.org)
	2 points by Jazgot 161 days ago \| hide \| past \| pdf \| 1 comment
1134.	LLMs Corrupt Your Documents When You Delegate (arxiv.org)
	4 points by achrono 162 days ago \| hide \| past \| pdf \| 2 comments
1135.	Memory in the Age of AI Agents (arxiv.org)
	2 points by fittingopposite 162 days ago \| hide \| past \| pdf \| 1 comment
1136.	Ouroboros: Dynamic Weight Generation for Recursive Transformers (arxiv.org)
	2 points by OsamaJaber 162 days ago \| hide \| past \| pdf \| discuss
1137.	LogAct: Enabling agentic reliability via shared logs (arxiv.org)
	2 points by pramodbiligiri 162 days ago \| hide \| past \| pdf \| discuss
1138.	Decoupled DiLoCo for Resilient Distributed Pre-Training (arxiv.org)
	3 points by matt_d 163 days ago \| hide \| past \| pdf \| discuss
1139.	The Geometry of Forgetting (arxiv.org)
	3 points by ashwing1984 163 days ago \| hide \| past \| pdf \| 1 comment
1140.	There Will Be a Scientific Theory of Deep Learning (arxiv.org)
	367 points by jamie-simon 163 days ago \| hide \| past \| pdf \| 167 comments
	More