New Links | ML News

ML Newsnew | past | best | rss | submit

961.	Hyperagents (Meta Research) (arxiv.org)
	2 points by galsapir 134 days ago \| hide \| past \| pdf \| discuss
962.	AI assistants can be hijacked and manipulated by inaudible sounds (arxiv.org)
	3 points by chbint 134 days ago \| hide \| past \| pdf \| discuss
963.	StreamIndex: Memory-bounded compressed sparse attention via streaming top-k (arxiv.org)
	4 points by OsamaJaber 134 days ago \| hide \| past \| pdf \| discuss
964.	Event Tensor: A Unified Abstraction for Compiling Dynamic Megakernel (arxiv.org)
	6 points by matt_d 135 days ago \| hide \| past \| pdf \| discuss
965.	Measuring Security Without Fooling Ourselves: Why Benchmarking Agents Is Hard (arxiv.org)
	3 points by Timofeibu 135 days ago \| hide \| past \| pdf \| discuss
966.	Domain-Camouflaged Injection Attacks Evade Detection in Multi-Agent LLM Systems (arxiv.org)
	38 points by sbulaev 135 days ago \| hide \| past \| pdf \| 4 comments
967.	An AI system to help scientists write expert-level empirical software (arxiv.org)
	2 points by cyco130 135 days ago \| hide \| past \| pdf \| discuss
968.	Is Capability a Liability? More Capable Language Models Make Worse Forecasts (arxiv.org)
	1 point by sbulaev 135 days ago \| hide \| past \| pdf \| discuss
969.	Meow-Omni 1: a multi-modal feline LLM (arxiv.org)
	3 points by Otterly99 135 days ago \| hide \| past \| pdf \| discuss
970.	Moss: Self-Evolution Through Source-Level Rewriting in Autonomous Agent Systems (arxiv.org)
	11 points by Timofeibu 135 days ago \| hide \| past \| pdf \| discuss
971.	Matching Principle: Adversarial, augmentation, etc. are estimators of one matrix (arxiv.org)
	2 points by visark_42 135 days ago \| hide \| past \| pdf \| discuss
972.	PopPy: Opportunistically Exploiting Parallelism in Python Compound AI Apps (arxiv.org)
	1 point by matt_d 135 days ago \| hide \| past \| pdf \| discuss
973.	CODA: Rewriting Transformer Blocks as GEMM-Epilogue Programs (arxiv.org)
	105 points by matt_d 135 days ago \| hide \| past \| pdf \| 12 comments
974.	Negation Neglect: When models fail to learn negations in training (arxiv.org)
	2 points by chr15m 135 days ago \| hide \| past \| pdf \| discuss
975.	Lecture Notes on Statistical Physics and Neural Networks (arxiv.org)
	3 points by Anon84 136 days ago \| hide \| past \| pdf \| discuss
976.	Multi-Stream LLMs: new paper on parallelizing/separating prompts, thinking, I/O (arxiv.org)
	156 points by atomicthumbs 136 days ago \| hide \| past \| pdf \| 16 comments
977.	DashAttention: Differentiable and Adaptable Sparse Hierarchical Attention (arxiv.org)
	9 points by cmogni1 136 days ago \| hide \| past \| pdf \| discuss
978.	PULSELoCo: 17x less trainer-to-trainer bandwidth in distributed RL post-training (arxiv.org)
	3 points by synapz_org 136 days ago \| hide \| past \| pdf \| discuss
979.	Tracking Capabilities for Safer Agents (arxiv.org)
	2 points by matt_d 137 days ago \| hide \| past \| pdf \| discuss
980.	Methodology for Selecting Runtime Architecture Patterns for LLM Agents (arxiv.org)
	2 points by Anon84 137 days ago \| hide \| past \| pdf \| discuss
981.	Sycophantic AI Decreases Prosocial Intentions and Promotes Dependence (2025) (arxiv.org)
	2 points by mpweiher 137 days ago \| hide \| past \| pdf \| discuss
982.	Scalable Packed Layouts for Vector-Length-Agnostic ML Code Generation (arxiv.org)
	2 points by matt_d 137 days ago \| hide \| past \| pdf \| discuss
983.	Stable Audio 3 (arxiv.org)
	99 points by guardienaveugle 137 days ago \| hide \| past \| pdf \| 18 comments
984.	Agent Security Is a Systems Problem (arxiv.org)
	5 points by sbulaev 137 days ago \| hide \| past \| pdf \| discuss
985.	Gram: Recursive reasoning models with stochastic latent trajectories (10M param) (arxiv.org)
	3 points by mrkn1 137 days ago \| hide \| past \| pdf \| 1 comment
986.	Code as Agent Harness (arxiv.org)
	3 points by s-xyz 137 days ago \| hide \| past \| pdf \| discuss
987.	SRM: Detecting slow-burn risk in AI-agent sessions before execution (arxiv.org)
	1 point by ilion_identity 137 days ago \| hide \| past \| pdf \| discuss
988.	WorldParticle: Unified World Simulation of Lagrangian Particles via Transformer (arxiv.org)
	1 point by vismit2000 137 days ago \| hide \| past \| pdf \| discuss
989.	RigidFormer: Learning Rigid Dynamics Using Transformers (arxiv.org)
	1 point by vismit2000 137 days ago \| hide \| past \| pdf \| discuss
990.	Positive Alignment: Artificial Intelligence for Human Flourishing (arxiv.org)
	1 point by lexandstuff 137 days ago \| hide \| past \| pdf \| discuss
	More