New Links | ML News

ML Newsnew | past | best | rss | submit

61.	AI and human tutoring yield equivalent GRE learning gains (arxiv.org)
	5 points by Betelbuddy 8 days ago \| hide \| past \| pdf \| 1 comment
62.	Self-Play Pretraining with Zero Data (arxiv.org)
	4 points by E-Reverance 8 days ago \| hide \| past \| pdf \| discuss
63.	Artificial Kuramoto Oscillatory Neurons (arxiv.org)
	1 point by jerlendds 8 days ago \| hide \| past \| pdf \| discuss
64.	Synthetic Hospital: Physician-Validated Longitudinal EHR Benchmark (arxiv.org)
	2 points by simonpure 8 days ago \| hide \| past \| pdf \| discuss
65.	Entropy-Based Guided Collaboration in Heterogeneous LLM Multi-Agent Systems (arxiv.org)
	1 point by wslh 8 days ago \| hide \| past \| pdf \| discuss
66.	Instrumental Monitor Evasion Emerges Under Ordinary Task Pressure (arxiv.org)
	2 points by sbulaev 8 days ago \| hide \| past \| pdf \| discuss
67.	Implicit Neural Representations with Periodic Activation Functions (2020) (arxiv.org)
	3 points by peter_d_sherman 8 days ago \| hide \| past \| pdf \| 1 comment
68.	Extracting and Characterizing Hidden Chain-of-Thought in Frontier Models (arxiv.org)
	2 points by jumploops 8 days ago \| hide \| past \| pdf \| discuss
69.	Control-Token Injection Suppresses Chain-of-Thought and Defeats Reasoning-Based (arxiv.org)
	1 point by sbulaev 8 days ago \| hide \| past \| pdf \| discuss
70.	Learning to Discover Interesting Mathematics (arxiv.org)
	1 point by E-Reverance 8 days ago \| hide \| past \| pdf \| discuss
71.	Scaling Laws for Neural Language Models (first "scaling laws" paper from 2020) (arxiv.org)
	2 points by thoughtpeddler 8 days ago \| hide \| past \| pdf \| discuss
72.	Study shows AI is as good as human tutoring for GRE learning gains (arxiv.org)
	6 points by cgn 8 days ago \| hide \| past \| pdf \| 1 comment
73.	Shutdown Sabotage Propensities in Multi-Agent Systems (arxiv.org)
	1 point by sbulaev 9 days ago \| hide \| past \| pdf \| discuss
74.	Semantics Delivery Network: Rethinking Web Retrieval for LLM Agents (arxiv.org)
	1 point by scottsiume 9 days ago \| hide \| past \| pdf \| 1 comment
75.	LensVLM: Selective Context Expansion for Compressed Visual Representation OfText (arxiv.org)
	1 point by KitN 9 days ago \| hide \| past \| pdf \| discuss
76.	Log-Depth Recurrent Language Modeling (arxiv.org)
	1 point by E-Reverance 9 days ago \| hide \| past \| pdf \| discuss
77.	Memory Control Signals Emerge Before Action in Long Horizon Agents (arxiv.org)
	2 points by simonpure 9 days ago \| hide \| past \| pdf \| discuss
78.	Greedy Decoding Is Not Precision-Invariant: Cross-Precision Output Divergence (arxiv.org)
	1 point by sbulaev 9 days ago \| hide \| past \| pdf \| discuss
79.	LatentPort: Cross-model recurrent state transfer without prefix replay (arxiv.org)
	1 point by mmprotest 10 days ago \| hide \| past \| pdf \| discuss
80.	CliffCompaction: Cost-Efficient Compaction for Long-Horizon Coding Agents (arxiv.org)
	1 point by simonpure 10 days ago \| hide \| past \| pdf \| discuss
81.	Double Descent and Malign Overfitting in Diffusion Models (arxiv.org)
	2 points by Betelbuddy 10 days ago \| hide \| past \| pdf \| discuss
82.	Code Critique: Intent, Drift, and Spotlight for AI-Generated Diffs at Scale (arxiv.org)
	2 points by raahelb 10 days ago \| hide \| past \| pdf \| discuss
83.	HySparse2: Hybrid Sparse Attention with Two-Level KV Sharing (arxiv.org)
	2 points by ksec 10 days ago \| hide \| past \| pdf \| discuss
84.	Recursive self-improvement of AI research agents (arxiv.org)
	3 points by handfuloflight 10 days ago \| hide \| past \| pdf \| 1 comment
85.	Training a Language Model End-to-End in Rust: An Experience Report (arxiv.org)
	2 points by Brajeshwar 10 days ago \| hide \| past \| pdf \| discuss
86.	PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from LLMs (arxiv.org)
	1 point by Bluestein 10 days ago \| hide \| past \| pdf \| discuss
87.	The Pain Axis: LLMs Represent Self-Directed Harm and Act to Relieve It (arxiv.org)
	1 point by simonpure 10 days ago \| hide \| past \| pdf \| discuss
88.	Testing AIs on 68 of the hardest open Erdos problems, verified in Lean (arxiv.org)
	3 points by tadamcz 10 days ago \| hide \| past \| pdf \| discuss
89.	Recursive self-improvement of AI research agents (arxiv.org)
	2 points by sbulaev 10 days ago \| hide \| past \| pdf \| discuss
90.	Specs cut defects in AI-generated code from 148 to 23 across five models (arxiv.org)
	1 point by sandeepdhuri 10 days ago \| hide \| past \| pdf \| discuss
	More