New Links | ML News

ML Newsnew | past | best | rss | submit

901.	Why Larger Models Learn More: Capacity, Interference, Rare-Task Retention (arxiv.org)
	3 points by matt_d 125 days ago \| hide \| past \| pdf \| discuss
902.	PassNet: Scaling Large Language Models for Graph Compiler Pass Generation (arxiv.org)
	2 points by matt_d 125 days ago \| hide \| past \| pdf \| discuss
903.	Memo: Memory as a Model (arxiv.org)
	2 points by melvinroest 125 days ago \| hide \| past \| pdf \| discuss
904.	Unlocking the Working Memory of Large Language Models for Latent Reasoning (arxiv.org)
	2 points by korbip 126 days ago \| hide \| past \| pdf \| discuss
905.	Enhancing Multi-Agent Communication Through Attention Steering (arxiv.org)
	4 points by ankitg12 126 days ago \| hide \| past \| pdf \| discuss
906.	Memory as Action: Autonomous Context Curation for Long-Horizon Agentic Tasks (arxiv.org)
	9 points by ankitg12 126 days ago \| hide \| past \| pdf \| discuss
907.	Scaling Laws Meet Model Architecture: Toward Inference-Efficient LLMs (arxiv.org)
	2 points by matt_d 126 days ago \| hide \| past \| pdf \| discuss
908.	Rotary GPU: Exploring Local Execution for Large MoE Models Under Limited VRAM (arxiv.org)
	41 points by dryarzeg 127 days ago \| hide \| past \| pdf \| 4 comments
909.	Autonomous LLM Agent Worms (arxiv.org)
	2 points by ankitg12 127 days ago \| hide \| past \| pdf \| discuss
910.	Scaling Laws for Agent Harnesses via Effective Feedback Compute (arxiv.org)
	1 point by veryluckyxyz 127 days ago \| hide \| past \| pdf \| discuss
911.	stable-worldmodel-v1: Reproducible World Modeling Research and Evaluation (arxiv.org)
	2 points by petethomas 128 days ago \| hide \| past \| pdf \| discuss
912.	Understanding Inference Scaling for LLMs: Bottlenecks, Trade-Offs, and Perf (arxiv.org)
	6 points by matt_d 128 days ago \| hide \| past \| pdf \| discuss
913.	AI Propaganda factories with language models (arxiv.org)
	6 points by rramadass 128 days ago \| hide \| past \| pdf \| 1 comment
914.	Cassandra: Enabling Reasoning LLMs at Edge via Self-Speculative Decoding (arxiv.org)
	4 points by chrsw 128 days ago \| hide \| past \| pdf \| discuss
915.	Negation Neglect: When models fail to learn negations in training (arxiv.org)
	3 points by johnbarron 128 days ago \| hide \| past \| pdf \| 2 comments
916.	StoryScope: Investigating Idiosyncrasies in AI Fiction (arxiv.org)
	1 point by ironyman 128 days ago \| hide \| past \| pdf \| discuss
917.	Continuous Diffusion Models Can Obey Formal Syntax (arxiv.org)
	2 points by matt_d 128 days ago \| hide \| past \| pdf \| discuss
918.	Can Go AIs be adversarially robust? (arxiv.org)
	1 point by Kotlopou 129 days ago \| hide \| past \| pdf \| discuss
919.	SIA: Self Improving AI with Harness and Weight Updates (arxiv.org)
	3 points by mitchwainer 129 days ago \| hide \| past \| pdf \| discuss
920.	Generative Recursive ReAsoning Models (Gram) (arxiv.org)
	7 points by ijidak 129 days ago \| hide \| past \| pdf \| discuss
921.	AutoScientists: Self-Organizing Agent Teams for Experimentation (arxiv.org)
	4 points by Anon84 129 days ago \| hide \| past \| pdf \| discuss
922.	Paris 2.0: Video diffusion model trained on decentralized, heterogeneous GPUs (arxiv.org)
	7 points by royychacker 129 days ago \| hide \| past \| pdf \| 1 comment
923.	Omissive Bias: Benchmarking LLM Answers to Ethical Decision-Making (arxiv.org)
	2 points by pseudolus 129 days ago \| hide \| past \| pdf \| discuss
924.	DeltaBox: Scaling Stateful AI Agents with Ms-Level Sandbox Checkpoint/Rollback (arxiv.org)
	2 points by fofoz 129 days ago \| hide \| past \| pdf \| discuss
925.	Gemini Embedding 2: A Native Multimodal Embedding Model from Gemini (arxiv.org)
	4 points by simonpure 129 days ago \| hide \| past \| pdf \| discuss
926.	Muse-Autoskill: Self-Evolving Agents via Skill Creation and Memory (arxiv.org)
	3 points by nilen 129 days ago \| hide \| past \| pdf \| discuss
927.	Harness Sensitivity Is Non-Monotone Across LLM Agent Tiers (arxiv.org)
	3 points by simonpure 129 days ago \| hide \| past \| pdf \| discuss
928.	Pimmur, can LLM simulate human collective behavior? (arxiv.org)
	2 points by xiaoluolyg 129 days ago \| hide \| past \| pdf \| discuss
929.	Agent Security Is a Systems Problem (arxiv.org)
	3 points by yakkomajuri 130 days ago \| hide \| past \| pdf \| discuss
930.	Agents Thinking Fast and Slow: A Talker-Reasoner Architecture (arxiv.org)
	3 points by jalcazar 130 days ago \| hide \| past \| pdf \| discuss
	More