연구ENArXiv AI
Hyperagents
arXiv:2603.19461v1 Announce Type: new Abstract: Selfimproving AI systems aim to reduce reliance on human engineering by learning to improve their own learning and problemsolving processes.
AI News
Stay up to date with the latest AI industry news.
연구ENArXiv AI
Teaching an Agent to Sketch One Part at a Time
arXiv:2603.19500v1 Announce Type: new Abstract: We develop a method for producing vector sketches one part at a time. To do this, we train a multimodal language modelbased agent using a novel multiturn processreward reinforcement learning following supervised finetuning.
연구ENArXiv AI
Learning to Disprove: Formal Counterexample Generation with Large Language Models
arXiv:2603.19514v1 Announce Type: new Abstract: Mathematical reasoning demands two critical, complementary skills: constructing rigorous proofs for true statements and discovering counterexamples that disprove false ones.
연구ENArXiv AI
ItinBench: Benchmarking Planning Across Multiple Cognitive Dimensions with Large Language Models
arXiv:2603.19515v1 Announce Type: new Abstract: Large language models LLMs with advanced cognitive capabilities are emerging as agents for various reasoning and planning tasks. Traditional evaluations often focus on specific reasoning or planning questions within controlled environments.
연구ENArXiv AI
PA2D-MORL: Pareto Ascent Directional Decomposition based Multi-Objective Reinforcement Learning
arXiv:2603.19579v1 Announce Type: new Abstract: Multiobjective reinforcement learning MORL provides an effective solution for decisionmaking problems involving conflicting objectives.
연구ENArXiv AI
PowerLens: Taming LLM Agents for Safe and Personalized Mobile Power Management
arXiv:2603.19584v1 Announce Type: new Abstract: Battery life remains a critical challenge for mobile devices, yet existing power management mechanisms rely on static rules or coarsegrained heuristics that ignore user activities and personal preferences.
연구ENArXiv AI
HyEvo: Self-Evolving Hybrid Agentic Workflows for Efficient Reasoning
arXiv:2603.19639v1 Announce Type: new Abstract: Although agentic workflows have demonstrated strong potential for solving complex tasks, existing automated generation methods remain inefficient and underperform, as they rely on predefined operator libraries and homogeneous LLMonly workflows in...
연구ENArXiv AI
A Subgoal-driven Framework for Improving Long-Horizon LLM Agents
arXiv:2603.19685v1 Announce Type: new Abstract: Large language model LLMbased agents have emerged as powerful autonomous controllers for digital environments, including mobile interfaces, operating systems, and web browsers.
연구ENArXiv AI
Stepwise: Neuro-Symbolic Proof Search for Automated Systems Verification
arXiv:2603.19715v1 Announce Type: new Abstract: Formal verification via interactive theorem proving is increasingly used to ensure the correctness of critical systems, yet constructing large proof scripts remains highly manual and limits scalability.
산업ENThe Verge AI
AI influencer awards season is upon us
First came the AI beauty pageant. Then the AI music contests. Now, there is an award for AI Personality of the Year perhaps the inevitable next step for the AI influencer economy as it fights for legitimacy amid an onslaught of AI slop and grifters and transforms from quirky novelty into a serious...
모델 출시ENOpenAI Blog
Creating with Sora Safely
To address the novel safety challenges posed by a stateoftheart video model as well as a new social creation platform, we’ve built Sora 2 and the Sora app with safety at the foundation. Our approach is anchored in concrete protections.
산업AI타임스
커서의 코딩 모델은 문샷 'K2.5' 미세조정 버전..."모델 재활용의 순환"
뛰어난 가성비로 주목받았던 커서의 새로운 모델이 문샷 AI의 오픈 웨이트 '키미 K2.5'를 미세조정한 버전으로 확인됐다.문샷은 21일현지시간 X를 통해 "커서의 컴포저 2Composer 2 출시를 축하한다"라며 "키미 K2.5가 그 기반이 된 것을 자랑스럽게 생각한다"라고 밝혔다. 이어 "커서의 지속적인 사전 학습 및 고성능 강화 학습을 통해 우리 모델이 효과적으로 통합되는 모습을 보니, 개방형 모델 생태계의 모범적인 사례라고 생각한다"라고 덧붙였다.이에 앞서 커서는 19일 자체 모델 컴포저 2를 공개하고 자체 플랫폼에 탑