Community feed
A focused stream of recent stories from the sources curated for this community. Latest: PXDepth: Pixel-Space Modeling for Structure Preserving Monocular Depth Estimation, Towards Safer RAG: Only Agents Capable of System 2 Thinking may Access Untrusted Documents, and There is No Theoretical Curse of Multilinguality For Embedding Space Structure. Page 313.
Planet AI Developer Filter
Abstract page for arXiv paper 2608.16984: PXDepth: Pixel-Space Modeling for Structure Preserving Monocular Depth Estimation
Planet AI Developer Filter
Abstract page for arXiv paper 2608.17153: Towards Safer RAG: Only Agents Capable of System 2 Thinking may Access Untrusted Documents
Planet AI Developer Filter
Abstract page for arXiv paper 2608.17088: There is No Theoretical Curse of Multilinguality For Embedding Space Structure
Planet AI Developer Filter
Abstract page for arXiv paper 2608.17051: Institution-Specific LLM Prompting Recovers PHI That De-identification Systems and Their Gold Standards Both Miss
Planet AI Developer Filter
Abstract page for arXiv paper 2608.17050: Cross-Model Memory Transfer via Target-Side Reader Adaptation
Entities: Target
Planet AI Developer Filter
Abstract page for arXiv paper 2608.16975: Margin-Regularized Structured Semantic Alignment for Brain-Language Correspondence
Planet AI Developer Filter
Abstract page for arXiv paper 2608.17124: A decodability criterion predicts when hidden-state selection beats majority voting in large language models
Planet AI Developer Filter
Abstract page for arXiv paper 2608.17071: KernelArc: A Multi-Agent Framework for GPU Kernel Optimization
Planet AI Developer Filter
Abstract page for arXiv paper 2608.17053: Memory Is Communication: The Frontier Between Remembering and Signaling
Planet AI Developer Filter
Abstract page for arXiv paper 2608.17007: SkillEffect: Checked Lowering for Memory-Bounded Agent Tools
Planet AI Developer Filter
Abstract page for arXiv paper 2608.16977: The Problem Is the Problem: Towards Scalable Mathematical Discovery
Planet AI Developer Filter
Abstract page for arXiv paper 2608.16956: The Price of Thinking: Reasoning Effort as a Model-Specific API Contract