[AI Paper] LLM-in-Sandbox Elicits General Agentic Intelligence
📄 LLM-in-Sandbox Elicits General Agentic Intelligence 개요 이 연구는 LLM-in-Sandbox라는 새로운 패러다임을 제안합니다. LLM에게 코드 샌드박스(가상 컴퓨터) 내에서 자유롭게 탐색할 수 있는 환경을 제공하여, 코딩이 아닌 일반 도메인에서도 일반 에이전트 지능(General Agentic Intelligence)을 발현시킬 수 있음을 입증합니다. [!tip] 핵심 발견 강력한 에이전트 LLM은 추가…
[AI Paper] CI4A: Semantic Component Interfaces for Agents Empowering Web Automation
📄 CI4A: Semantic Component Interfaces for Agents Empowering Web Automation 개요 대규모 언어 모델(LLM)이 고수준 의미론적 계획(semantic planning)에서는 탁월한 성과를 보여주지만, 세밀한 저수준 웹 컴포넌트 조작(fine-grained, low-level web component manipulations)에서는 여전히 제한적이다. 이 논문은 에이전트를 위해 최적화된 상호작용…
[AI Paper] An Agentic Operationalization of DISARM for FIMI Investigation on Social Media
📄 An Agentic Operationalization of DISARM for FIMI Investigation on Social Media 📌 1단계: 기본 정보 제목 An Agentic Operationalization of DISARM for FIMI Investigation on Social Media (에이전트 기반 DISARM 운영화: 소셜미디어 외국 정보 조작 및 개입 조사) 저자 Kevin Tseng, Juan Carlos…
[AI Paper] From Passive Metric to Active Signal: The Evolving Role of Uncertainty Quantification in Large Language Models
From Passive Metric to Active Signal: The Evolving Role of Uncertainty Quantification in Large Language Models 📌 1단계: 기본 정보 제목 From Passive Metric to Active Signal: The Evolving Role of Uncertainty Quantification in Large Language Models 저자 Jiaxin Zhang…
[AI Paper] Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification
📄 Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification 1단계: 기본 정보 논문 정보 제목: Inference-Time Scaling of Verification: Self-Evolving Deep Research Agents via Test-Time Rubric-Guided Verification…
[AI Paper] Zero-shot Adaptable Task Planning for Autonomous Construction Robots
Zero-shot Adaptable Task Planning for Autonomous Construction Robots 메타 정보 항목 내용 저자 Hossein Naderi, Alireza Shojaei, Lifu Huang, Philip Agee, Kereshmeh Afsari, Abiola Akanmu 소속 Virginia Tech (Myers-Lawson School of Construction) arXiv ID 2601.14091v1 제출일…
[AI Paper] Where Do AI Coding Agents Fail? An Empirical Study of Failed Agentic Pull Requests in GitHub
Where Do AI Coding Agents Fail? An Empirical Study of Failed Agentic Pull Requests in GitHub 📌 1단계: 기본 정보 항목 내용 제목 Where Do AI Coding Agents Fail? An Empirical Study of Failed Agentic Pull Requests in GitHub 저자 Ramtin Ehsani, Sakshi Pathak, Shriya Rawal,…
[AI Paper] VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning
VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning 📌 1단계: 기본 정보 제목: VideoThinker: Building Agentic VideoLLMs with LLM-Guided Tool Reasoning 저자: Chenglin Li (Zhejiang University, Shanghai Innovation Institute) Qianglong Chen (Zhejiang…
[AI Paper] TransportAgents: a multi-agents LLM framework for traffic accident severity prediction
TransportAgents: a multi-agents LLM framework for traffic accident severity prediction 📌 1단계: 기본 정보 항목 내용 제목 TransportAgents: a multi-agents LLM framework for traffic accident severity prediction 저자 Zhichao Yang¹, Jiashu He², Jinxuan Fan³, Cirillo…
[AI Paper] 📄 Towards Reliable ML Feature Engineering via Planning in Constrained-Topology of LLM Agents
📄 Towards Reliable ML Feature Engineering via Planning in Constrained-Topology of LLM Agents 메타 정보 저자: Himanshu Thakur, Anusha Kamath, Anurag Muthyala, Dhwani Sanmukhani, Smruthi Mukund, Jay Katukuri 출처: arXiv:2601.10820v1 발표일: 2026년 1월 라이선스: CC BY 4.0…