As NLP evaluation shifts from static benchmarks to multi-turn interactive settings, LLM-based simulators have become widely used as user proxies, serving two roles: generating user turns and providing evaluation signals. Yet, these simulations are frequently assumed to be faithful to real human behaviors, often without rigorous verification. We formalize the Sim2Real gap in user simulation and present the first study running the full $τ$-bench protocol with real humans (451 participants, 165 tasks), benchmarking 31 LLM simulators across proprietary, open-source, and specialized families using the User-Sim Index (USI), a metric we introduce to quantify how well LLM simulators resemble real user interactive behaviors and feedback. Behaviorally, LLM simulators are excessively cooperative, stylistically uniform, and lack realistic frustration or ambiguity, creating an "easy mode" that inflates agent success rates above the human baseline. In evaluations, real humans provide nuanced judgments across eight quality dimensions while simulated users produce uniformly more positive feedback; rule-based rewards are failing to capture rich feedback signals generated by human users. Overall, higher general model capability does not necessarily yield more faithful user simulation. These findings highlight the importance of human validation when using LLM-based user simulators in the agent development cycle and motivate improved models for user simulation.
Paradigms of Intelligence Team, Google（谷歌智能范式团队）；University of Chicago（芝加哥大学）；Santa Fe Institute（圣菲研究所）；Antikythera, Berggruen Institute（安提基特拉与伯格格林研究所）；University of California, San Diego（加州大学圣地亚哥分校）
Mind Meets Space: Rethinking Agentic Spatial Intelligence from a Neuroscience-inspired Perspective
School of EEE, Nanyang Technological University（南洋理工大学电子工程系）；Department of Civil Engineering, Tsinghua University（清华大学土木工程系）；Dr. B. R. Ambedkar National Institute of Technology, Jalandhar（B.R.阿姆贝卡尔国家理工学院，贾兰德赫）；KTH Royal Institute of Technology（皇家理工学院）；College of AI, Tsinghua University（清华大学人工智能学院）；CCDS, Nanyang Technological University（南洋理工大学CCDS）
Being Considerate as a Pathway Towards Pluralistic Alignment for Agentic AI
University of Toronto（多伦多大学）；Vector Institute（向量研究所）；Schwartz Reisman Institute for Technology and Society（施瓦茨·赖斯曼技术与社会研究所）；Pontificia Universidad Católica de Chile（智利天主教大学）；Centro Nacional de Inteligencia Artificial（国家人工智能中心）
Gap the (Theory of) Mind: Sharing Beliefs About Teammates' Goals Boosts Collaboration Perception, Not Performance
Faculty of Data & Decisions Science Technion - Israel Institute of Technology（数据与决策科学学院，技术学院-以色列理工学院）；Department of Computer Science Tufts University（计算机科学系，塔夫茨大学）
Bridging the Cognitive Gap: A Unified Memory Paradigm for 6G Agentic AI-RAN
Sun Yat-sen University, China（中山大学，中国）；Zhejiang University, China（浙江大学，中国）；University of Exeter, U.K.（埃克塞特大学，英国）；Singapore University of Technology and Design, Singapore（新加坡科技设计大学，新加坡）；Purple Mountain Laboratories, Nanjing, China（紫金山实验室，南京，中国）
奖励函数是强化学习（RL）中必不可少的，作为激励智能体解决给定任务的指导信号，然而，也是众所周知的难以设计。在许多情况下，只有不完美的奖励是可用的，这对RL代理造成了重大的性能损失。在这项研究中，我们提出了一个统一的离线策略优化方法，\textit{RGM (Reward Gap Minimization)}，它可以巧妙地处理不同类型的不完美奖励。RGM被制定为一个双层优化问题：上层优化的奖励校正项，执行访问分布匹配w.r.t.一些专家数据;较低层用校正的奖励解决悲观RL问题。通过利用较低层的二元性，我们得到了一个易于处理的算法，使基于采样的学习没有任何在线交互。综合实验表明，RGM实现了优越的性能，现有的方法在不同的设置不完美的奖励。此外，RGM可以有效地纠正错误的或不一致的奖励对专家的偏好和检索有用的信息，从有偏见的奖励。
Mind the Gap: How Elicitation Protocols Shape the Stated-Revealed Preference Gap in Language Models
University of Oxford（牛津大学）；Max Planck Institute for Biological Cybernetics（生物信息学Max Planck研究所）；University of Tuebingen（图宾根大学）；Cardiff University（卡迪夫大学）；Cambridge–Boston Alignment Initiative (CBAI)（剑桥-波士顿对齐倡议）
Mind the Gap: Bridging the Divide Between AI Aspirations and the Reality of Autonomous Characterization
National Renewable Energy Laboratory（国家可再生能源实验室）；University of Cincinnati（俄亥俄大学）；School of Materials Engineering（材料工程学院）；Purdue University（普渡大学）；School of Mechanical Engineering（机械工程学院）；Pacific Northwest National Laboratory（太平洋西北国家实验室）；Energy and Environment Directorate（能源与环境管理局）；Colorado School of Mines（科罗拉多矿业学院）；Renewable and Sustainable Energy Institute（可再生能源与可持续能源研究所）；University of Colorado Boulder（科罗拉多大学博尔德分校）
Mind the Gap: Conformative Decoding to Improve Output Diversity of Instruction-Tuned Large Language Models
School of Computing University of Kent（肯特大学计算机学院）；Leiden Institute of Advanced Computer Science Universiteit Leiden（莱顿先进计算机科学研究所）；Cheriton School of Computer Science University of Waterloo（滑铁卢大学查里顿计算机科学学院）
Mind the Gap: Promoting Missing Modality Brain Tumor Segmentation with Alignment
School of Robotics Xi’an Jiaotong-Liverpool University Suzhou, China；School of Advanced Technology Xi'an Jiaotong-Liverpool University Suzhou, China；School of Advanced Technology Xi’an Jiaotong-Liverpool University Suzhou, China；Data Science Research Center Duke Kunshan University Suzhou, China
