연구ENArXiv AI
Continually self-improving AI
arXiv:2603.18073v1 Announce Type: new Abstract: Modern language modelbased AI systems are remarkably powerful, yet their capabilities remain fundamentally capped by their human creators in three key ways.
AI News
Stay up to date with the latest AI industry news.
연구ENArXiv AI
Multi-Trait Subspace Steering to Reveal the Dark Side of Human-AI Interaction
arXiv:2603.18085v1 Announce Type: new Abstract: Recent incidents have highlighted alarming cases where humanAI interactions led to negative psychological outcomes, including mental health crises and even user harm.
연구ENArXiv AI
Adaptive Domain Models: Bayesian Evolution, Warm Rotation, and Principled Training for Geometric and Neuromorphic AI
arXiv:2603.18104v1 Announce Type: new Abstract: Prevailing AI training infrastructure assumes reversemode automatic differentiation over IEEE754 arithmetic.
연구ENArXiv AI
Don't Vibe Code, Do Skele-Code: Interactive No-Code Notebooks for Subject Matter Experts to Build Lower-Cost Agentic Workflows
arXiv:2603.18122v1 Announce Type: new Abstract: SkeleCode is a naturallanguage and graphbased interface for building workflows with AI agents, designed especially for less or nontechnical users.
연구ENArXiv AI
Efficient Dense Crowd Trajectory Prediction Via Dynamic Clustering
arXiv:2603.18166v1 Announce Type: new Abstract: Crowd trajectory prediction plays a crucial role in public safety and management, where it can help prevent disasters such as stampedes.
연구ENArXiv AI
TeachingCoach: A Fine-Tuned Scaffolding Chatbot for Instructional Guidance to Instructors
arXiv:2603.18189v1 Announce Type: new Abstract: Higher education instructors often lack timely and pedagogically grounded support, as scalable instructional guidance remains limited and existing tools rely on generic chatbot advice or nonscalable teaching center humanhuman consultations.
연구ENArXiv AI
Access Controlled Website Interaction for Agentic AI with Delegated Critical Tasks
arXiv:2603.18197v1 Announce Type: new Abstract: Recent studies reveal gaps in delegating critical tasks to agentic AI that accesses websites on the user's behalf, primarily due to limited access control mechanisms on websites designed for agentic AI.
연구ENArXiv AI
A Computationally Efficient Learning of Artificial Intelligence System Reliability Considering Error Propagation
arXiv:2603.18201v1 Announce Type: new Abstract: Artificial Intelligence AI systems are increasingly prominent in emerging smart cities, yet their reliability remains a critical concern.
연구ENArXiv AI
Retrieval-Augmented LLM Agents: Learning to Learn from Experience
arXiv:2603.18272v1 Announce Type: new Abstract: While large language models LLMs have advanced the development of generalpurpose agents, achieving robust generalization to unseen tasks remains a significant challenge.
산업ENThe Verge AI
OpenAI is planning a desktop ‘superapp’
OpenAI is working on a desktop "superapp" that merges its ChatGPT app, the Codex AI coding app, and its AIpowered Atlas browser into one app, The Wall Street Journal reports.
산업인공지능신문
美 해군, ‘데이터 엔진’ 실전 배치…“함정이 스스로 학습하는 플랫폼으로 진화”
물리적 AI의 미래를 선도하는 글로벌 기업 어플라이드 인튜이션Applied Intuition이 미국 해군U.S. Navy에 대규모 데이터 엔진 프로그램을 공급하며, 인공지능AI 기반 전장 체계 전환이 본격화되고 있다.19일현지시간 어플라이드 인튜이션은 美 해군의 첫 대규모 데이터 엔진인 ‘데이터 엣지 컬렉션 키트Data Edge Collection Kit. 이하, DECK’를 성공적으로 인도했다고 발표했다. DECK는 해군의 로보틱 및 자율 시스템 프로그램인 PAE RASPortfolio Acquisition E
산업ENTechCrunch AI
Jeff Bezos reportedly wants $100 billion to buy and transform old manufacturing firms with AI
The Amazon magnate has a new project centered around acquiring industrial firms and revamping them with AI technology.