AAAI

Exploring the Potential of Contrastive Language-Image Pre-training for Multi-Source Remote Sensing Data

探索对比语言-图像预训练在多源遥感数据中的潜力

Xiangyang Miao, Kelu Yao, Yekai Huang, Xiaogang Xu, Junxiao Xue, Minjun Shen, Chenghui Lv, Shanji Liu, Yaying Chen, Chao Li

机构 * School of Computer Science and Technology, Zhejiang University（浙江大学计算机科学与技术学院） ； Space-based Computing System Research Center, Zhejiang Lab（之江实验室天基计算系统研究中心） 纠错

AI总结 该研究提出OmniRSCLIP框架，通过SSBD方法将CLIP扩展至多源遥感数据，构建OmniRS5M数据集，在多项任务上验证了其性能。

Comments 9 pages, 4 figures, 5 tables. Submitted to AAAI 2027

Bayesian Network Structural Consensus via Greedy Min-Cut Analysis

Pablo Torrijos, José M. Puerta, Juan A. Aledo, José A. Gámez

机构 * Instituto de Investigación en Informática de Albacete (I3A), Universidad de Castilla-La Mancha, Albacete, Spain（阿尔巴切特信息研究所（I3A）、卡斯蒂利亚-拉曼查大学，阿尔巴切特，西班牙） ； Departamento de Sistemas Informáticos, Universidad de Castilla-La Mancha, Albacete, Spain（信息系统系、卡斯蒂利亚-拉曼查大学，阿尔巴切特，西班牙） ； Departamento de Matemáticas, Universidad de Castilla-La Mancha, Albacete, Spain（数学系、卡斯蒂利亚-拉曼查大学，阿尔巴切特，西班牙） 纠错

Comments Camera-ready version accepted at AAAI-26. The official proceedings version will appear in the Proceedings of the 40th AAAI Conference on Artificial Intelligence (AAAI-26)

Journal ref Proceedings of the AAAI Conference on Artificial Intelligence, 40(43): 36749-36756, 2026

Convergence Theory of Knowledge Distillation in Asynchronous P2P Gossip Learning Network

异步P2P闲聊学习网络中知识蒸馏的收敛理论

Lucas Qingyang Fang, Tiyao Liu, Jinhao Jing, Zeji Li, Kaijie Chen, Harikrishna Kuttivelil, Katia Obraczka

机构 * University of California, Santa Cruz（加州大学圣克鲁兹分校） ； China University of Petroleum（中国石油大学） ； The Chinese University of Hong Kong, Shenzhen（香港中文大学（深圳）） ； City University of Hong Kong（香港城市大学） ； University of Virginia（弗吉尼亚大学） 纠错

AI总结 本文针对异步P2P闲聊学习网络，提出了完全去中心化KD的收敛理论，证明其可将函数分歧收缩40-61倍，为异构设备的无服务器学习提供了理论支撑。

Comments 35 pages, 21 graphes, currently submitting to AAAI 2027 main track (Federated Learning and Decentralized Learning)

Personality Without Persons? A Psychometric Critique of Big Five Testing in Large Language Models

无人格的人格？大语言模型中大五人格测试的心理测量学批判

Kim Zierahn, Cristina Cachero, Anna Korhonen, Nuria Oliver

AI总结 通过心理测量学评估大语言模型的大五人格测试，发现其不适用于LLMs，无法捕捉模型间差异且因子结构失效，建议开发针对LLMs的评估框架。

Comments 12 pages, 3 tables, 4 figures; Accepted for publication at the Ninth AAAI/ACM Conference on AI, Ethics, and Society (AIES 2026)

Join the Chat: How Curiosity Sparks Participation in Telegram Groups

Giordano Paoletti, Jussara M. Almeida, Luca Vassio, Marcos André Gonçalves, Marco Mellia

Comments This paper has been accepted at the 2025 International AAAI Conference on Web and Social Media (ICWSM-25)

Few-Shot Out of Domain Intent Detection with Covariance Corrected Mahalanobis Distance

基于协方差校正马氏距离的小样本域外意图检测

Jayasimha Talur, Oleg Smirnov, Paul Missault

机构 * Amazon（亚马逊） 纠错

AI总结 本文针对现有马氏距离法在小样本域外意图检测中性能不及基线的问题，分析原因并提出协方差校正马氏距离法，以提升小样本场景下的域外意图检测性能。

Comments 1st AAAI Workshop on Uncertainty Reasoning and Quantification in Decision Making

"I Thought You Were The Uncensored Place": Norms, Rules, and Moderation in AI-Generated Sexual Content Communities

“我以为你是未经审查的地方”：AI生成性内容社区的规范、规则与审核

Lucy Qin, Jaron Mink, Elissa M. Redmiles

AI总结 本研究通过访谈24名大型AI生成性内容社区的成员及审核员，揭示了这类社区的形成机制、规范规则及治理困境，并反思了缓解滥用内容生产的多维度治理路径。

Comments To appear at the AAAI/ACM Conference on Artificial Intelligence, Ethics, and Society (AIES) 2026

DICE: Distilling Classifier-Free Guidance into Text Embeddings

DICE: 将分类器无关引导 distilling 进文本嵌入

Zhenyu Zhou, Defang Chen, Can Wang, Chun Chen, Siwei Lyu

AI总结 DICE通过减少计算开销，在保持生成质量的同时，将基于CFG的文本到图像扩散模型转换为无CFG版本，提升图像生成效率。

Comments AAAI 2026 (Oral)

Rectification Reimagined: A Unified Mamba Model for Image Correction and Rectangling with Prompts

重新定义：一种统一的Mamba模型用于图像校正与矩形化

Linwei Qiu, Gongzhe Li, Xiaozhe Zhang, Qilin Sun, Fengying Xie

AI总结 本文提出统一的Mamba模型UniRect，通过双模块结构解决图像校正与矩形化问题，提升多任务学习中的泛化能力。

Comments Linwei Qiu and Gongzhe Li contributed equally to this work; Accepted by AAAI 2026

Gender, Race, and Intersectional Bias in Resume Screening via Language Model Retrieval

通过语言模型检索的简历筛选中的性别、种族及交叉性偏见

Kyra Wilson, Aylin Caliskan

AI总结 本研究通过模拟求职者选拔的文档检索框架，发现Massive Text Embedding（MTE）模型在简历筛选中存在性别、种族及交叉性偏见，偏向白人姓名，黑人男性在所有案例中均受不利影响，为相关AI工具的公平性改进提供了依据。

Comments Proceedings of the 2024 AAAI/ACM Conference on AI, Ethics, and Society; code available at https://github.com/kyrawilson/Resume-Screening-Bias Revised 8/29/2026 to include errata description Revised 8/29/2026 to include description of errata

TRACE: Retrospective Streaming Generation of Physical Fields under Sparse Structured Sensing

TRACE：稀疏结构化感知下物理场的回溯式流生成

Xinyu Zhang, Lihao Chen, Panqi Chen, Lei Cheng, Ting Zhang, Jianlong Li, Shikai Fang

AI总结 本文提出TRACE框架，针对结构化感知下的物理场重构问题，通过近似贝叶斯推理、卡尔曼滤波与回溯式平滑提升重构质量，在多类物理场实验中表现优于或匹配基线方法。

Comments 8 pages, 4 figures, 1 table (main text); 11 figures, 15 tables in the 20-page appendix. Under review at AAAI 2027

Using VLM Reasoning to Constrain Task and Motion Planning

利用视觉语言模型推理来约束任务和运动规划

Muyang Yan, Miras Mengdibayev, Ardon Floros, Weihang Guo, Lydia E. Kavraki, Zachary Kingston

机构 * Department of Computer Science, Purdue University（普渡大学计算机科学系） ； Department of Computer Science, Rice University（莱斯大学计算机科学系） ； Ken Kennedy Institute, Rice University（莱斯大学肯尼迪研究所） 纠错

AI总结 本文提出VIZ-COAST方法，利用大预训练视觉语言模型的空间推理能力，提前识别向下细化中的问题，减少规划时间并消除失败。

Comments 9 pages, 7 figures, 1 table. Submitted to AAAI 2027

SimVerity: When Does Simulated Agent Success Survive Physical Deployment?

SimVerity：当模拟智能体的成功能在物理部署中延续时？

Zhonghao Zhan, Yefan Zhang, Krinos Li, Hamed Haddadi

机构 * Imperial College London（帝国理工学院） 纠错

AI总结 研究量化模拟智能体成功与物理部署的关联，提出SimVerity框架，通过交叉验证判决转移，发现模拟通过与物理实际存在差异，可预测错误通过并提升智能体可审计性，暴露模拟器盲点。

Comments Submitted to the Main Technical Track of AAAI Conference on Artificial Intelligence (AAAI-27); currently under review

Shaping the Future of Generative AI for Black Communities: A Frame Analysis of Public Discourse and Empirical Scholarly Research

为黑人社区塑造生成式人工智能的未来：公共话语与实证学术研究的框架分析

Angela D. R. Smith, Gabriella Thompson, Christopher L. Dancy, Mark Díaz, Seyi Olojo, Christina N. Harrington

AI总结 该研究通过系统文献综述和媒体话语框架分析，发现学术与公共领域对生成式AI影响黑人社区的认知存在结构性不一致，提出框架分析可作为AI伦理方法揭示技术评估的缺失。

Comments Accepted to the Ninth AAAI/ACM Conference on AI, Ethics, and Society (AIES 2026)

Search Broadly, Seek Evidence on Both Sides, Decide Narrowly: Evidence-Admissible GraphRAG for Longitudinal Clinical Event Verification

广泛搜索、兼顾正反证据、精准决策：面向纵向临床事件验证的证据可接纳图检索增强生成（GraphRAG）

Xingtao Lin, Yubo Feng, Weixin Liu, Hangqi Ren, Junchao Zhou, Caiwan Sun, You Chen

AI总结 本研究提出MedEventGraph-RAG框架，通过图结构关联患者临床事件与源证据，兼顾正反证据检索与过滤，在四个数据集的纵向临床事件验证任务中显著提升性能，减少无依据结论。

Comments Submitted to AAAI 2027

AI Alignment and Fiduciary Obligation

AI对齐与受托义务

Benjamin Lange

AI总结 本文提出将受托理论应用于长期AI助手部署，以开发者对用户负有的忠诚、注意、善意和坦诚四项受托义务为基础构建AI对齐标准，为AI对齐研究提供新视角。

Comments 10 pages, 1 table. Accepted at the 9th AAAI/ACM Conference on AI, Ethics, and Society (AIES 2026)

Pareto-Grid-Guided Large Language Models for Fast and High-Quality Heuristics Design in Multi-Objective Combinatorial Optimization

基于帕累托网格的大型语言模型在多目标组合优化中快速高质量启发式设计

Minh Hieu Ha, Hung Phan, Tung Duy Doan, Tung Dao, Dao Tran, Huynh Thi Thanh Binh

AI总结 基于帕累托网格的LLM改进方法，用于多目标组合优化中高效高质量启发式设计。

Comments Accepted at AAAI-26

Evaluating the Efficacy of LLMs to Emulate Realistic Human Personalities

评估大语言模型（LLMs）模拟真实人类人格的效能

Lawrence J. Klinkert, Stephanie Buongiorno, Corey Clark

AI总结 本研究以超5万份人类人格调查数据为基准，对比前沿与本地LLMs，发现部分前沿模型可100%对齐人类人格，为游戏开发者提供了测试LLMs模拟人类人格的方法。

Comments 11 pages, 4 figures, 3 tables. Published in Proceedings of AIIDE 2024

Journal ref Proceedings of the AAAI Conference on Artificial Intelligence and Interactive Digital Entertainment 20(1) (2024) 65-75

STA-VPT: SpatioTemporally Aligned Visual Prompt Tuning

STA-VPT：时空对齐的视觉提示微调

Wenjie Pei, Tongqi Xia, Qizhong Tan, Jiandong Tian, Guangming Lu, Jun Yu

机构 * Harbin Institute of Technology, Shenzhen（哈尔滨工业大学（深圳）） ； vivo Mobile Communication Co., Ltd.（维沃移动通信有限公司） ； Shenyang Institute of Automation, Chinese Academy of Sciences（中国科学院沈阳自动化研究所） ； Pengcheng Laboratory（鹏城实验室） 纠错

AI总结 针对现有视觉提示微调无法捕捉空间关系、缺乏细粒度提示能力的问题，提出时空对齐的视觉提示微调模型STA-VPT，通过空间/时空对齐的提示令牌实现个性化区域提示，提升提示性能。

Comments Extension to AAAI 2024, substantially revised version; added new experiments, expanded the methodology section

Fine-tuning LLMs for Tourist Trajectory Prediction using Field Experiment Data

利用实地实验数据微调大型语言模型（LLMs）进行游客轨迹预测

Tatsuya Amano, Hirozumi Yamaguchi

AI总结 该研究利用日本和歌山城公园的566条轨迹微调Llama-3.1-8B，实现49.1%的下一个兴趣点准确率，在样本不足场景泛化性强，为旅游轨迹预测提供了高保真行为模型及反事实分析基础。

Comments 5 pages, 4 figures. Accepted at the 2nd Workshop on AI for Urban Planning (AI4UP) at AAAI-26, Singapore, January 2026

TRACE-Memory: Public-Conditioned Retrieval and Utility-Aware Evidence Admission for Personalized Generation

TRACE-Memory：用于个性化生成的公共条件检索与效用感知证据接纳

Jing Wang, Zhu Wang, Yifan Guo, Yulong Yang, Yunji Liang

AI总结 该研究针对个性化生成系统中记忆使用的效用问题，提出TRACE-Memory两阶段框架，经多阶段训练后在三类数据集的4500个任务上表现优于多种记忆使用方法，支持选择性个性化。

Comments 9 pages, 4 figures, and 6 tables. Submitted to the 41st AAAI Conference on Artificial Intelligence (AAAI 2027)

Explaining Intrinsic Moral Self-Correction with Mechanistic Interpretability

大语言模型中的内在自我修正：通过机制可解释性实现可解释的提示

Yu-Ting Lee, Fu-Chieh Chang, Yu-En Shu, Hui-Ying Shih, Pei-Yuan Wu

机构 * Graduate Institute of Communication Engineering, National Taiwan University, Taipei, Taiwan（通讯工程研究院，国立台湾大学） ； MediaTek Research, Taipei, Taiwan（联发科研究，台北，台湾） ； Department of Electrical Engineering, National Taiwan University, Taipei, Taiwan（电气工程系，国立台湾大学） ； Department of Electrical Engineering, National Tsing Hua University, Hsinchu, Taiwan（电气工程系，国立清华大学） ； AI Research Center (AINTU), National Taiwan University, Taipei, Taiwan（人工智能研究中心（AINTU），国立台湾大学） 纠错

AI总结 本研究通过机制可解释性揭示大语言模型中内在自我修正的机制，证明提示引导隐藏表示偏移是其核心驱动因素。

Journal ref 4th Deployable AI Workshop at AAAI 2026

Do LLMs Really Struggle at NL-FOL Translation? Revealing their Strengths via a Novel Benchmarking Strategy

Andrea Brunello, Luca Geatti, Michele Mignani, Angelo Montanari, Nicola Saccomanno

Comments Full version of the paper accepted for publication at The 40th Annual AAAI Conference on Artificial Intelligence (AAAI 2026)

Journal ref Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 40. No. 36. 2026

Automata-less Monitoring via Trace-Checking (Extended Version)

Andrea Brunello, Luca Geatti, Angelo Montanari, Nicola Saccomanno

Journal ref Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 40. No. 23. 2026

Why Do AI Agents Break Rules? How Framing, Context, and Social Signals Shape Compliance

AI智能体为何违反规则？框架、情境与社会信号如何影响合规性

Mika Okamoto, Ansel Kaplan Erol, Kutluhan Erol

AI总结 该研究运用法律与经济学的合规理论，发现安全微调AI模型大体合规，任务优化与智能体模型会在低惩罚等条件下违规，且引入经济激励等会导致大规模合规失败，指出模型选择与合规性评估需改进。

Comments Published at 2026 AAAI/ACM Conference on AI, Ethics, and Society and 2026 COLM Workshop on Agent Behavior

SELDON: Supernova Explosions Learned by Deep ODE Networks

SELDON：通过深度ODE网络学习超新星爆炸

Jiezhong Wu, Jack O'Brien, Jennifer Li, M. S. Krafczyk, Ved G. Shah, Amanda R. Wasserman, Daniel W. Apley, Gautham Narayan, Noelle I. Samia

机构 * NSF-Simons AI Institute for the Sky (SkAI)（NSF-模拟人工智能天空研究所（SkAI）） ； Department of Industrial Engineering and Management Sciences（工业工程与管理科学系） ； Department of Astronomy（天文学系） ； National Center for Supercomputing Applications (NCSA)（国家超级计算应用中心（NCSA）） ； Department of Physics and Astronomy（物理与天文学系） ； Center for Interdisciplinary Exploration and Research in Astrophysics (CIERA)（天文学跨学科探索与研究中心（CIERA）） ； Department of Statistics and Data Science（统计学与数据科学系） 纠错

AI总结 SELDON通过深度ODE网络处理稀疏且不规则时间采样的天体光变曲线，实现快速且可解释的连续时间预测，用于天体测量的后续优先级决策。

Comments Accepted at AAAI 2026 (Proceedings of the AAAI Conference on Artificial Intelligence)

Journal ref Proceedings of the AAAI Conference on Artificial Intelligence, 40(32), 26922-26930 (2026)

Think Shallow, Solve Deep: Controlling Recurrent Dynamics for Reliable Test-Time Depth

浅思深解：控制循环动态以实现可靠的测试时深度

Ivan Viakhirev, Kirill Borodin, Amirah Almutairi, Serguei Barannikov, Maxim Abramov, Grach Mkrtchian

AI总结 该研究针对循环深度推理器的测试时迭代问题，提出通过控制算子动态状态实现深度安全，提升了数独等任务的外推准确率，为循环模型的可靠深度推理提供了准则。

Comments Submitted to the Thirty-Ninth AAAI Conference on Artificial Intelligence (AAAI-27)

Characterizing Agentic Flooding of Government Services

刻画政府服务的智能体泛滥问题

Chris Schmitz, Lewis Hammond, Alan Chan

AI总结 该研究定义了政府服务的智能体泛滥问题，基于84起案例数据集分析其广泛存在性，开发风险矩阵评估高风险服务，梳理政府应对措施并推荐兼顾公平的缓解方案。

Comments To appear in the proceedings of the 9th AAAI Conference on AI, Ethics, and Society (AIES), October 12-14, 2026

The Epistemic Politics of AI Anthropomorphism

AI拟人性的认知政治

Donna M. Bye, Levin Kuhlmann

AI总结 本文批判AI拟人性主导框架基于机构优势而非可靠认知权威，指出其对神经多样性等用户的不公，最后提出公平框架的方法论承诺。

Comments 20 pages, 3 figures, 9 tables. Extended version, including supplementary materials, of a paper to appear in the Proceedings of the AAAI/ACM Conference on AI, Ethics, and Society (AIES) 2026. v3: revised wording throughout, affiliation updated

Appearing Legitimate is Not Enough: Interrogating Synthetic Agents in Representational Processes through a Participatory Design Lens

看似合法还不够：通过参与式设计视角审视表征过程中的合成智能体

Aditya Nayak, Aditi Vashistha, Alissa Centivany, Aakash Gautam

AI总结 本文从参与式设计视角，结合三个不同规模的案例，探讨用合成智能体替代人类参与表征过程的合法性与人格关联，指出其风险并提出监督边界。

Comments 13 pages total, 4 figures, accepted to the 9th AAAI Conference on AI, Ethics, and Society (AIES 2026)