Large language model (LLM) agents face a structural tension: cloud agents provide strong reasoning but expose user data, while on-device agents preserve privacy at the cost of overall capability. Existing device-cloud designs treat this boundary as a compute split rather than a trust boundary suited to agentic workloads, and existing sanitizers force a choice between policy flexibility and the structural fidelity tool calls require. In this work, we develop PAAC, a privacy-aware agentic framework that aligns planner--executor decomposition with the device-cloud boundary so that role specialization itself becomes the privacy mechanism. The cloud agent reasons over typed placeholder tokens that preserve each sensitive value's reasoning role while discarding its content, while the on-device agent identifies sensitive spans and distills each step's execution outcome into compact key findings. Sanitization confines the on-device LLM to proposing which spans to mask, while a deterministic registry performs all substitution and reversal, keeping actions directly executable on device. On three agentic benchmarks under strict privacy settings, PAAC dominates the Pareto frontier of privacy and accuracy, improving average accuracy by 15-36\% and reducing average leakage by 2-6$\times$ over state-of-the-art device-cloud baselines, with the largest margins on privacy targets outside fixed entity taxonomies. We find consistent improvements on 17 additional benchmarks spanning 10 domains, including math, science, and finance.
Multi-agent Architecture Search via Agentic Supernet
National University of Singapore(新加坡国立大学);Tongji University(同济大学);University of Science and Technology of China(中国科学技术大学);Nanyang Technological University(南洋理工大学);Shanghai AI Laboratory(上海人工智能实验室)
Leiden University Leiden Netherlands;Leiden University \& AI Lab, Pegasystems Leiden Netherlands;Leiden University;Leiden University \& AI Lab, Pegasystems
CollabBench: Benchmarking and Unleashing Collaborative Ability of LLMs with Diverse Players via Proactive Engagement
Shanghai Institute of AI for Education(上海人工智能教育研究院);School of Computer Science(计算机科学学院);East China Normal University(东华大学);Tencent Inc.(腾讯公司);Shanghai Innovation Institute(上海创新研究院)
Rank-and-Reason: Multi-Agent Collaboration Accelerates Zero-Shot Protein Mutation Prediction
Shanghai Jiao Tong University(上海交通大学);Shanghai Innovation Institute(上海创新研究院);Southern University of Science(南方大学(科学));East China University of Science and Technology(东中国科学与技术大学)
Embracing Trustworthy Brain-Agent Collaboration as Paradigm Extension for Intelligent Assistive Technologies
University of Illinois Chicago(伊利诺伊大学芝加哥分校);MBZUAI(马克斯·普朗克人工智能研究所);McGill University(麦吉尔大学);The Chinese University of Hong Kong(香港中文大学);Nanyang Technological University(南洋理工大学);Tsinghua University(清华大学)
A Survey of Frontiers in LLM Reasoning: Inference Scaling, Learning to Reason, and Agentic Systems
Salesforce AI Research(Salesforce AI研究);National University of Singapore(国立新加坡大学);Nanyang Technological University(南洋理工大学);The Hong Kong University of Science and Technology (Guangzhou)(香港科学与技术大学(广州));A*STAR, Singapore(新加坡A*STAR)
Knowledge-Driven Agentic Scientific Corpus Distillation Framework for Biomedical Large Language Models Training
Computer Network Information Center, Chinese Academy of Sciences, Beijing, China(中国科学院计算机网络信息中心);University of Chinese Academy of Sciences, Beijing, China(中国科学院大学);Duke-NUS Medical School, National University of Singapore, Singapore(新加坡国立大学杜克-新加坡医学部)
Engineering Signals of Human-AI Collaboration in the Agentic Coding Era: A Longitudinal Analysis of 33,228 Pull Requests from vLLM and SGLang with Implications for Biomedical AI Agents and Bioinformatics Pipeline Developmen
Shanghai Jiao Tong University(上海交通大学);The Hong Kong University of Science and Technology(香港科技大学);University of Liverpool(利物浦大学);University of California Berkeley(加州大学伯克利分校);Shanghai Innovation Institute(上海创新研究院);University of California Davis(加州大学戴维斯分校);Virginia Tech(弗吉尼亚理工大学);University College London(伦敦大学学院)