Submissions from arxiv.org

		Context-Aware Membership Inference Attacks Against Pre-Trained LLMs (arxiv.org)
		2 points by felineflock 8 days ago \| past \| discuss
		Fast and Accurate Long Text Generation with Few-Step Diffusion Language Models (arxiv.org)
		4 points by gok 9 days ago \| past \| 1 comment
		Understanding RL for model training, and future directions with GRAPE (arxiv.org)
		33 points by sonabinu 9 days ago \| past \| 1 comment
		LLM probabilities cannot distinguish between possible and impossible language (arxiv.org)
		2 points by foobarqux 9 days ago \| past \| discuss
		Qwen3-Omni: first multimodal model with SoTA text, image, audio, and video perf (arxiv.org)
		2 points by walterbell 9 days ago \| past \| discuss
		Using LLMs to create datasets: reconstructing the historical memory of Colombia (arxiv.org)
		1 point by PaulHoule 9 days ago \| past \| discuss
		Do LLM Modules Generalize? A Study on Motion Generation for Autonomous Driving (arxiv.org)
		1 point by PaulHoule 10 days ago \| past \| discuss
		Can LIGO Detect Daylight Savings Time? (arxiv.org)
		9 points by belter 10 days ago \| past \| 1 comment
		Bit is all we need: binary normalized neural networks (arxiv.org)
		101 points by PaulHoule 10 days ago \| past \| 54 comments
		Federation of Agents: Semantics-Aware, Large-Scale Communication Fabric (arxiv.org)
		3 points by simonpure 10 days ago \| past \| discuss
		The Memory Paradox: Why Our Brains Need Knowledge in an Age of AI (arxiv.org)
		4 points by rahimnathwani 10 days ago \| past \| 1 comment
		Hierarchical Retrieval: The Geometry and a Pretrain-Finetune Recipe (arxiv.org)
		1 point by JnBrymn 10 days ago \| past \| discuss
		TimeCopilot: Framework for Forecasting combining Time Series Models with LLMs (arxiv.org)
		2 points by favoboa 10 days ago \| past \| discuss
		SimpleFold: Folding Proteins Is Simpler Than You Think (arxiv.org)
		2 points by gok 10 days ago \| past \| discuss
		GraphMend: Code Transformations for Fixing Graph Breaks in PyTorch 2 (arxiv.org)
		3 points by matt_d 10 days ago \| past \| discuss
		A Software Engineering Analysis of the XZ Utils Supply Chain Attack (arxiv.org)
		5 points by PaulHoule 10 days ago \| past \| 1 comment
		Just-in-time and distributed task representations in language models (arxiv.org)
		1 point by PaulHoule 10 days ago \| past \| discuss
		The Illusion of Readiness: Stress Testing Frontier Models on Medical Benchmarks (arxiv.org)
		6 points by mellosouls 10 days ago \| past \| discuss
		Report on the 63rd Annual International Mathematical Olympiad (arxiv.org)
		1 point by bikenaga 10 days ago \| past \| discuss
		A fast, strong, topologically meaningful and fun knot invariant (arxiv.org)
		52 points by bikenaga 10 days ago \| past \| 7 comments
		Quantized LLMss in Biomedical Natural Language Processing (arxiv.org)
		1 point by PaulHoule 10 days ago \| past \| discuss
		Ransomware 3.0: Self-Composing and LLM-Orchestrated (arxiv.org)
		1 point by PaulHoule 10 days ago \| past \| discuss
		Multi-Modal vs. Text-Based: Benchmarking LLM Strategies for Invoice Processing (arxiv.org)
		1 point by PaulHoule 11 days ago \| past \| discuss
		LIMI: Less Is More for Agency (arxiv.org)
		1 point by pella 11 days ago \| past \| discuss
		Design, analysis, and manufacturing of microstructured blade-like geometries (arxiv.org)
		2 points by PaulHoule 11 days ago \| past \| discuss
		Fill probability estimates in institutional bond trading with quantum computers (arxiv.org)
		2 points by polrjoy 11 days ago \| past \| 2 comments
		Weak Memory Model Formalisms: Introduction and Survey (arxiv.org)
		2 points by matt_d 11 days ago \| past \| discuss
		Why Language Models Hallucinate (arxiv.org)
		1 point by ummonk 11 days ago \| past \| discuss
		GPU Implementation of Second-Order Linear and Nonlinear Programming Solvers (arxiv.org)
		1 point by adgjlsfhk1 11 days ago \| past \| 1 comment
		Bluffing in Scrabble (arxiv.org)
		8 points by fanf2 11 days ago \| past \| discuss
		More