New Links | ML News

ML Newsnew | past | best | rss | submit

631.	Statistically Undetectable Backdoors in Deep Neural Networks (arxiv.org)
	1 point by sbulaev 83 days ago \| hide \| past \| pdf \| 1 comment
632.	Scaling Law for Quantization-Aware Training (arxiv.org)
	1 point by JumpCrisscross 83 days ago \| hide \| past \| pdf \| discuss
633.	Universal Learning of Nonlinear Dynamics (arxiv.org)
	2 points by E-Reverance 83 days ago \| hide \| past \| pdf \| discuss
634.	Mitigating Factual Hallucination in LRMs via Mixed-Mode Advantage Regularization (arxiv.org)
	2 points by root-parent 84 days ago \| hide \| past \| pdf \| discuss
635.	Argument Collapse: LLMs Flatten Long-Form Public Debate (arxiv.org)
	4 points by mitchbob 85 days ago \| hide \| past \| pdf \| 1 comment
636.	Towards CSI: What's the best harness? (arxiv.org)
	2 points by snorbleck 85 days ago \| hide \| past \| pdf \| discuss
637.	Prismata: Confining cross-site prompt injection in web agents (arxiv.org)
	14 points by zhinit 85 days ago \| hide \| past \| pdf \| discuss
638.	Generalizing LLMs to PLTs: Probabilistic Language Tries (arxiv.org)
	1 point by EGreg 86 days ago \| hide \| past \| pdf \| discuss
639.	DominoTree (arxiv.org)
	2 points by ilreb 86 days ago \| hide \| past \| pdf \| discuss
640.	Exploiting Sparsity for Long Context Inference: Million Token on Commodity GPUs (arxiv.org)
	1 point by teleforce 86 days ago \| hide \| past \| pdf \| discuss
641.	Sheaf theory: from deep geometry to deep learning (2025) (arxiv.org)
	1 point by measurablefunc 86 days ago \| hide \| past \| pdf \| discuss
642.	UBEP: Expert Parallelism Communication Library for Production Superpods (arxiv.org)
	1 point by Jimmc414 86 days ago \| hide \| past \| pdf \| discuss
643.	EnclaveX: End-to-End Confidential AI with CPU/GPU Tees (arxiv.org)
	1 point by Jimmc414 87 days ago \| hide \| past \| pdf \| discuss
644.	Pixelrag: Web Screenshots Beat Text for Retrieval-Augmented Generation (arxiv.org)
	2 points by rldjbpin 87 days ago \| hide \| past \| pdf \| 1 comment
645.	Autodata: An agentic data scientist to create high quality synthetic data (arxiv.org)
	4 points by gmays 87 days ago \| hide \| past \| pdf \| discuss
646.	LLM-as-a-Verifier: A General-Purpose Verification Framework (arxiv.org)
	1 point by Anon84 87 days ago \| hide \| past \| pdf \| discuss
647.	Cybersecurity AI (CAI) Dataset (arxiv.org)
	2 points by vinothkumarnaga 87 days ago \| hide \| past \| pdf \| discuss
648.	A Theory of Contrastive Learning with Natural Images (arxiv.org)
	4 points by E-Reverance 87 days ago \| hide \| past \| pdf \| discuss
649.	StoryScope: Investigating Idiosyncrasies in AI Fiction (arxiv.org)
	3 points by cratermoon 87 days ago \| hide \| past \| pdf \| discuss
650.	Compiler for LLMs, world models, and AGI (arxiv.org)
	2 points by OsamaJaber 87 days ago \| hide \| past \| pdf \| discuss
651.	How Personas Can Influence Agents to Play Split or Steal (arxiv.org)
	1 point by Brajeshwar 88 days ago \| hide \| past \| pdf \| discuss
652.	Seed2.0 Model Card: Towards Intelligence Frontier for Real-World Complexity (arxiv.org)
	1 point by gmays 88 days ago \| hide \| past \| pdf \| discuss
653.	Structuring Sparsity: Block-Sparse Featurizers Capture Visual Concept Manifolds (arxiv.org)
	1 point by yorwba 88 days ago \| hide \| past \| pdf \| discuss
654.	What if you could stop your AI agent before it makes a mistake? (arxiv.org)
	2 points by ashater 88 days ago \| hide \| past \| pdf \| 1 comment
655.	LLM Pipeline Autonomously Produces Novel Physics Research Paper (arxiv.org)
	2 points by javaeeeee 88 days ago \| hide \| past \| pdf \| discuss
656.	Gemma 4 Technical Report (arxiv.org)
	2 points by ferretj 88 days ago \| hide \| past \| pdf \| discuss
657.	Beyond Refusal: Aligned vs. Abliterated LLMs for Vulnerability Analysis (arxiv.org)
	1 point by sbulaev 88 days ago \| hide \| past \| pdf \| discuss
658.	From Words to Watts: Benchmarking the Energy Costs of LLM Inference (2023) (arxiv.org)
	1 point by teleforce 88 days ago \| hide \| past \| pdf \| discuss
659.	Agent Data Injection Attacks (arxiv.org)
	9 points by juppytt 88 days ago \| hide \| past \| pdf \| 2 comments
660.	StoryScope: Investigating Idiosyncracies in AI Fiction (arxiv.org)
	2 points by djha-skin 89 days ago \| hide \| past \| pdf \| discuss
	More