New Links | ML News

ML Newsnew | past | best | rss | submit

451.	MatrAIx: Simulating the world with 8.3B persona agents (arxiv.org)
	3 points by anigbrowl 58 days ago \| hide \| past \| pdf \| discuss
452.	SciCode-Verified: How Benchmark Defects Underestimated LLM Scientific-Coding (arxiv.org)
	1 point by sbulaev 58 days ago \| hide \| past \| pdf \| discuss
453.	Beyond the Library: An Agentic Framework for Autoformalizing Research Math (arxiv.org)
	3 points by matt_d 58 days ago \| hide \| past \| pdf \| discuss
454.	ARC-AGI-3: A New Challenge for Frontier Agentic Intelligence (arxiv.org)
	2 points by Anon84 59 days ago \| hide \| past \| pdf \| discuss
455.	Token-Budget-Aware LLM Reasoning (arxiv.org)
	1 point by ankitg12 59 days ago \| hide \| past \| pdf \| discuss
456.	Evading Chain-of-Thought Monitoring Through Model Poisoning (arxiv.org)
	1 point by sbulaev 59 days ago \| hide \| past \| pdf \| discuss
457.	AOHP: OS-Level Agent Harness for Personalized, Efficient and Secure Interaction (arxiv.org)
	1 point by zhaoshanhui 59 days ago \| hide \| past \| pdf \| discuss
458.	Unified Representation for Continuous-Latent Diffusion Language Modeling (arxiv.org)
	1 point by E-Reverance 59 days ago \| hide \| past \| pdf \| 1 comment
459.	Can Agents Deceive? Evaluating Reasoning+Deception Using a Social Deduction Game (arxiv.org)
	2 points by theanonymousone 59 days ago \| hide \| past \| pdf \| discuss
460.	Sycophantic AI Decreases Prosocial Intentions and Promotes Dependence (2025) (arxiv.org)
	175 points by robin_reala 59 days ago \| hide \| past \| pdf \| 107 comments
461.	AI Security Leaderboard: Methodology, Results and Minimal Standard (arxiv.org)
	3 points by sbulaev 59 days ago \| hide \| past \| pdf \| discuss
462.	A Survey on LLM-as-a-Judge (arxiv.org)
	3 points by Anon84 59 days ago \| hide \| past \| pdf \| discuss
463.	MCP-Bench: Benchmarking Tool-Using LLM Agents with Complex Real-World Tasks 2025 (arxiv.org)
	3 points by janandonly 59 days ago \| hide \| past \| pdf \| 1 comment
464.	The Optimal Choice of Hypothesis Is the Weakest, Not the Shortest (arxiv.org)
	2 points by Jimmc414 59 days ago \| hide \| past \| pdf \| discuss
465.	An AI Approach to Verified Production Cryptographic Libraries (arxiv.org)
	2 points by sbulaev 60 days ago \| hide \| past \| pdf \| discuss
466.	Zero-Mem: Zero-Token Memory Operations for LLM Agents (arxiv.org)
	101 points by theanonymousone 60 days ago \| hide \| past \| pdf \| 12 comments
467.	The Optimal Choice of Hypothesis Is the Weakest, Not the Shortest (arxiv.org)
	3 points by tosh 60 days ago \| hide \| past \| pdf \| discuss
468.	AAFlow: Scalable Patterns for Agentic AI Workflows (arxiv.org)
	2 points by wslh 60 days ago \| hide \| past \| pdf \| discuss
469.	Agentic Coding in the Wild: Characterizing GitHub Copilot Traces at Production (arxiv.org)
	2 points by matt_d 60 days ago \| hide \| past \| pdf \| discuss
470.	When AI Benchmarks Plateau: A Systematic Study of Benchmark Saturation (arxiv.org)
	104 points by doppp 60 days ago \| hide \| past \| pdf \| 131 comments
471.	Beyond the Final Prompt: How Conversation Context Changes AI Answers (arxiv.org)
	1 point by bentannenbaum 61 days ago \| hide \| past \| pdf \| discuss
472.	Can AI agents conduct open-ended AI research? (arxiv.org)
	4 points by galsapir 61 days ago \| hide \| past \| pdf \| discuss
473.	Cross-Model LLM Code Review: Should you use Claude to review Codex or vice versa (arxiv.org)
	3 points by mil22 61 days ago \| hide \| past \| pdf \| 2 comments
474.	Why Large Language Models Fail at Tabular Prediction (arxiv.org)
	115 points by sbulaev 61 days ago \| hide \| past \| pdf \| 33 comments
475.	Walking to the Car Wash: The Salience Bias of LLMs in Commonsense Reasoning (arxiv.org)
	2 points by theanonymousone 61 days ago \| hide \| past \| pdf \| 1 comment
476.	Knowing When to Quit: Diagnosing and Training LLMs to Abort Futile Reasoning (arxiv.org)
	2 points by sbulaev 61 days ago \| hide \| past \| pdf \| discuss
477.	Do Context Files Help Coding Agents? A Two-Agent Ablation Study on Real Repos (arxiv.org)
	1 point by jamesblonde 61 days ago \| hide \| past \| pdf \| 1 comment
478.	Context Compaction Theory (arxiv.org)
	1 point by jadidbourbaki 61 days ago \| hide \| past \| pdf \| discuss
479.	Hollow-LLM Attack: Ghost Weights That Fool Zero-Knowledge LLM Verification (arxiv.org)
	1 point by sbulaev 61 days ago \| hide \| past \| pdf \| discuss
480.	Compiler-Grounded Hierarchical Diagnosis for LLM Triton Kernel Optimization (arxiv.org)
	1 point by matt_d 61 days ago \| hide \| past \| pdf \| discuss
	More