From AI Assistant to AI Scientist: Autonomous Discovery of LLM-RL Algorithms with LLM Agents
Shanghai Key Laboratory of Data Science, School of Computer Science, Fudan University(上海数据科学关键实验室,计算机科学学院,复旦大学);School of Data Science, Fudan University(数据科学学院,复旦大学)
Discovering improved policy optimization algorithms for language models remains a costly manual process requiring repeated mechanism-level modification and validation. Unlike simple combinatorial code search, this problem requires searching over algorithmic mechanisms tightly coupled with training dynamics while reusing empirical evidence across iterations. We propose POISE, a closed-loop framework for automated discovery of policy optimization algorithms for language models. POISE maintains a structured, genealogically linked archive linking proposals, executable implementations, standardized evaluations, and natural-language reflections to support evidence-driven iteration. In mathematical reasoning experiments starting from GRPO, POISE evaluates 64 candidate algorithms and discovers improved mechanisms, including analytic-variance scaling and validity masking. The best variant improves weighted Overall from 47.8 to 52.5 (+4.6) and increases AIME25 pass@32 from 26.7% to 43.3%, demonstrating the feasibility of automated policy optimization discovery while supporting interpretable design principles.
Autonomous Workflow for Multimodal Fine-Grained Training Assistants Towards Mixed Reality
Centrum Wiskunde & Informatica(数学与计算机科学中心);King Abdullah University of Science and Technology(阿卜杜拉国王科技大学);University College London(伦敦大学学院);Shandong University(山东大学);TU Delft(代尔夫特理工大学)
AI Urban Scientist: Multi-Agent Collaborative Automation for Urban Research
Vanke School of Public Health, Tsinghua University, Beijing, China(范克学校公共卫生学院,清华大学,北京,中国);Zhongguancun Academy, Beijing, China(中关村学院,北京,中国);Department of Electronic Engineering, Tsinghua University, Beijing, China(电子工程系,清华大学,北京,中国)
Measuring Data Science Automation: A Survey of Evaluation Tools for AI Assistants and Agents
Leverhulme Centre for the Future of Intelligence, University of Cambridge, UK(未来智能研究中心、剑桥大学);Valencian Research Institute for Artificial Intelligence (VRAIN), Universitat Politècnica de València, Spain(瓦伦西亚人工智能研究 institutes (VRAIN)、瓦伦西亚理工大学)
Risks of AI Scientists: Prioritizing Safeguarding Over Autonomy
Yale University(耶鲁大学);National Library of Medicine, National Institutes of Health(国家医学图书馆,国立卫生研究院);Mila-Quebec AI Institute(魁北克AI研究所);Shanghai Jiao Tong University(上海交通大学);OPPO Research Institute(OPPO研究院);Reichman University(里奇曼大学)
Department of Machine Learning, MBZUAI, Abu Dhabi, UAE(机器学习系,MBZUAI,阿布扎比,阿联酋);Department of Natural Language Processing, MBZUAI, Abu Dhabi, UAE(自然语言处理系,MBZUAI,阿布扎比,阿联酋)
ROMA: Real-time Omni-Multimodal Assistant with Interactive Streaming Understanding
CAS Key Laboratory of AI Safety, Institute of Computing Technology, CAS, Beijing, China(中国科学院人工智能安全重点实验室,计算技术研究所,中国科学院,北京,中国);University of Chinese Academy of Sciences, Beijing, China(中国科学院大学,北京,中国);Tsinghua University, Beijing, China(清华大学,北京,中国)
Are Large Language Models Reliable AI Scientists? Assessing Reverse-Engineering of Black-Box Systems
Department of Computer Science Princeton University(计算机科学系普林斯顿大学);Department of Computer Science Department of Psychology Princeton University(计算机科学系心理学系普林斯顿大学)
Scaling Laws in Scientific Discovery with AI and Robot Scientists
University of Toronto(多伦多大学);Istituto Italiano di Tecnologia(意大利技术研究院);Universita di Genova(热那亚大学);Tsinghua University(清华大学);Zhejiang University(浙江大学);Rutgers University(罗格斯大学);Harvard University(哈佛大学);Georgia Tech(佐治亚理工学院);University College of London(伦敦大学学院)
SpectraQuery: A Hybrid Retrieval-Augmented Conversational Assistant for Battery Science
Mechanical Engineering, Stanford University(斯坦福大学机械工程系);Applied Energy Division, SLAC National Accelerator Laboratory(SLAC国家加速器实验室应用能源部门);Institute of Computational and Mathematical Engineering, Stanford University(斯坦福大学计算与数学工程研究所)