AAAI

SecMoE: Communication-Efficient Secure MoE Inference via Select-Then-Compute

SecMoE: 通过选择-然后计算实现通信高效的安全MoE推理

Bowen Shen, Yuyue Chen, Peng Yang, Bin Zhang, Xi Zhang, Zoe L. Jiang

AI总结 SecMoE通过选择-然后计算方法，在保持隐私的同时实现高效安全的MoE推理，使模型规模扩大63倍，运行时间仅增加15.2倍。

Comments Accepted by AAAI 2026

Robust Evacuation for Multi-Drone Failure in Drone Light Shows

多无人机故障下的鲁棒疏散

Minhyuk Park, Aloysius K. Mok, Tsz-Chiu Au

AI总结 本研究提出了一种针对多无人机故障的鲁棒疏散算法，通过预测故障无人机轨迹并利用隐藏无人机实现快速恢复，提高无人机灯光秀的可靠性。

Comments Accepted to AAAI-26 Bridge Program B10: Making Embodied AI Reliable with Testing and Formal Verification

N2N-GQA: Noise-to-Narrative for Graph-Based Table-Text Question Answering Using LLMs

N2N-GQA：基于图的表格-文本问答中的噪声到叙事使用大语言模型

Mohamed Sharafath, Aravindh Annamalai, Ganesh Murugan, Aravindakumar Venugopalan

机构 * comcast India Engineering Center (CIEC)（comcast印度工程中心） 纠错

AI总结 N2N-GQA通过构建动态证据图提升基于图的表格-文本问答性能，实现零样本多跳问答的高准确率。

Comments Accepted at an AAAI 2026 Workshop

Improving Day-Ahead Grid Carbon Intensity Forecasting by Joint Modeling of Local-Temporal and Cross-Variable Dependencies Across Different Frequencies

通过联合建模本地时间依赖性和跨变量依赖性提升日提前电网碳强度预测

Bowen Zhang, Hongda Tian, Adam Berry, A. Craig Roussac

AI总结 本文提出了一种联合建模本地时间依赖性和跨变量依赖性的新型模型，用于提升电网碳强度预测的准确性。

Comments 2026 40th AAAI Conference on Artificial Intelligence

A Unified Shape-Aware Foundation Model for Time Series Classification

面向时间序列分类的统一形状感知基础模型

Zhen Liu, Yucheng Wang, Boyuan Li, Junhao Zheng, Emadeldeen Eldele, Min Wu, Qianli Ma

AI总结 UniShape是一种面向时间序列分类的统一形状感知基础模型，通过自适应聚合多尺度判别子序列提升模型可解释性，并在多个数据集上实现最先进的分类性能。

Comments Accepted in AAAI 2026

Representing Sounds as Neural Amplitude Fields: A Benchmark of Coordinate-MLPs and A Fourier Kolmogorov-Arnold Framework

将声音表示为神经振幅场：坐标MLP的基准测试及傅里叶科尔莫戈洛夫-阿诺德框架

Linfei Li, Lin Zhang, Zhong Wang, Fengyi Zhang, Zelin Li, Ying Shen

AI总结 本文提出傅里叶ASR框架，利用傅里叶级数和科尔莫戈洛夫-阿诺德定理，通过傅里叶-KAN网络高效表示音频信号，提升音频表示的鲁棒性和质量。

Comments Accepted by AAAI 2025. Code: https://github.com/lif314/Fourier-ASR

Human-in-the-Loop Interactive Report Generation for Chronic Disease Adherence

具有临床医生的交互式报告生成用于慢性病依从性

Xiaotian Zhang, Jinhong Yu, Pengwei Yan, Le Jiang, Xingyi Shen, Mumo Cheng, Xiaozhong Liu

AI总结 本文提出一种具有临床医生的交互式报告生成系统，通过AI生成与医生审查的分工，提高慢性病依从性报告的效率和准确性，同时确保问责制。

Comments 5 pages, 3 figures. Accepted at the AAAI 2026 Workshop on AI for Healthy Aging and Longevity

PCoKG: Personality-aware Commonsense Reasoning with Debate

PCoKG：具有辩论机制的个性感知常识推理

Weijie Li, Zhongqing Wang, Guodong Zhou

AI总结 PCoKG通过引入辩论机制构建个性感知常识知识图谱，提升对话生成的一致性与个性化水平。

Comments Accept by AAAI-2026

COVR:Collaborative Optimization of VLMs and RL Agent for Visual-Based Control

COVR：视觉控制中视觉语言模型与强化学习代理的协同优化

Canming Xia, Peixi Peng, Guang Tan, Zhan Su, Haoran Xu, Zhenxian Liu, Luntong Li

AI总结 COVR通过协同优化视觉语言模型与强化学习代理，利用RL生成数据提升VLM语义推理能力，并通过动作先验引导策略学习，实现视觉控制任务中的高效优化。

Comments The paper was accepted by the Fortieth AAAI Conference on Artificial Intelligence (AAAI-26)

TEAS: Trusted Educational AI Standard: A Framework for Verifiable, Stable, Auditable, and Pedagogically Sound Learning Systems

TEAS: 可信教育AI标准：可验证、稳定、可审计和教学上可靠的系统框架

Abu Syed

机构 * Abu Syed（阿布·赛德） 纠错

AI总结 本文提出TEAS框架，通过可验证性、稳定性、可审计性和教学合理性四个支柱，为教育AI系统提供可信度标准，强调系统架构而非模型能力的重要性。

Comments 19 pages, 6 tables, accepted at AAAI-26 (DAI Workshop) in Singapore

SmartSplat: Feature-Smart Gaussians for Scalable Compression of Ultra-High-Resolution Images

SmartSplat: 基于特征的智能高斯用于超高清图像的可扩展压缩

Linfei Li, Lin Zhang, Zhong Wang, Ying Shen

AI总结 SmartSplat通过特征感知的高斯方法实现超高清图像的高效压缩，兼顾压缩比与重建质量，展现强可扩展性。

Comments Accepted by AAAI 2026

Bidirectional Channel-selective Semantic Interaction for Semi-Supervised Medical Segmentation

双向通道选择性语义交互用于半监督医学分割

Kaiwen Huang, Yizhe Zhang, Yi Zhou, Tianyang Xu, Tao Zhou

AI总结 本文提出双向通道选择性语义交互框架，通过语义-空间扰动、通道选择性路由器和双向通道交互提升半监督医学图像分割性能。

Comments Accepted to AAAI 2026. Code at: https://github.com/taozh2017/BCSI

SPAN: Benchmarking and Improving Cross-Calendar Temporal Reasoning of Large Language Models

SPAN：大型语言模型跨日历时间推理的基准测试与改进

Zhongjian Miao, Hao Fu, Chen Wei

机构 * Li Auto（利亚特） 纠错

AI总结 SPAN通过开发时间代理提升大型语言模型跨日历时间推理能力，实验显示其准确率高达95.31%

Comments Accepted at the AAAI 2026 conference. This version includes the supplementary appendix

Symbolic Planning and Multi-Agent Path Finding in Extremely Dense Environments with Unassigned Agents

符号规划与在极度密集环境中无分配代理的多智能体路径寻找

Bo Fu, Zhe Chen, Rahul Chandan, Alex Barbosa, Michael Caldara, Joey Durham, Federico Pecora

AI总结 本文提出五种基于搜索的算法，用于解决在极度密集环境中无分配代理的多智能体路径寻找问题，通过图搜索方法高效生成重新排列计划。

Comments AAAI Conference on Artificial Intelligence (AAAI-26)

CliCARE: Grounding Large Language Models in Clinical Guidelines for Decision Support over Longitudinal Cancer Electronic Health Records

CliCARE: 在纵向癌症电子健康记录上基于临床指南 grounding 大型语言模型以支持决策

Dongchen Li, Jitao Liang, Wei Li, Xiaoyu Wang, Longbing Cao, Kun Yu

AI总结 CliCARE 通过将纵向癌症 EHRs 转换为时序知识图谱并结合临床指南，为肿瘤科医生提供证据支持的决策支持。

Comments Accepted in AAAI Conference on Artificial Intelligence (AAAI-26, Oral)

Branch, or Layer? Zeroth-Order Optimization for Continual Learning of Vision-Language Models

分支还是层？零阶优化用于视觉-语言模型的持续学习

Ziwei Liu, Borui Kang, Wei Li, Hangjie Yuan, Yanbing Yang, Wenbin Li, Yifan Zhu, Tao Feng, Jun Luo

AI总结 本文提出一种基于零阶优化的视觉-语言模型持续学习方法，通过模态感知策略提升模型逃避局部极小值的能力，实验表明在四个基准测试中取得最佳性能。

Comments Published in the Proceedings of the AAAI Conference on Artificial Intelligence (AAAI 2026)

Detect All-Type Deepfake Audio: Wavelet Prompt Tuning for Enhanced Auditory Perception

检测所有类型的深度伪造音频：用于增强听觉感知的小波提示微调

Yuankun Xie, Ruibo Fu, Zhiyong Wang, Xiaopeng Wang, Songjun Cao, Long Ma, Haonan Cheng, Long Ye

AI总结 本文提出小波提示微调方法，用于增强所有类型深度伪造音频检测性能，通过自监督学习优化模型参数，实现更高效的跨类型音频检测。

Comments Accepted to AAAI 2026

Toward Maturity-Based Certification of Embodied AI: Quantifying Trustworthiness Through Measurement Mechanisms

迈向基于成熟度的具身AI认证：通过测量机制量化可信度

Michael C. Darling, Alan H. Hesu, Michael A. Mardikes, Brian C. McGuigan, Reed M. Milewicz

机构 * Michael C. Darling ； Alan H. Hesu ； Michael A. Mardikes ； Brian C. McGuigan ； Reed M. Milewicz 纠错

AI总结 本文提出基于成熟度的具身AI认证框架，通过量化机制评估可信度，并通过无人机检测案例验证其可行性。

Comments Accepted to AAAI-26 Bridge Program B10: Making Embodied AI Reliable with Testing and Formal Verification

From Policy to Logic for Efficient and Interpretable Coverage Assessment

从策略到逻辑：为高效和可解释的覆盖评估而设

Rhitabrat Pokharel, Hamid Reza Hassanzadeh, Ameeta Agrawal

AI总结 本文提出了一种结合覆盖意识检索器和符号规则推理的方法，以提高医疗覆盖政策审查的效率和可解释性，同时降低模型成本并提升评估准确性。

Comments Accepted at AIMedHealth @ AAAI 2026

Automated Reproducibility Has a Problem Statement Problem

自动化可重复性存在一个问题陈述问题

Thijs Snelleman, Peter Lundestad Lawrence, Holger H. Hoos, Odd Erik Gundersen

AI总结 本文提出了一种通用的问题陈述框架，用于自动化经验性AI研究的可重复性，通过自动提取研究假设、实验和解释，验证了其在不同子领域中的有效性。

Comments Accepted at RAI Workshop @ AAAI 2026

Attribute-Aware Controlled Product Generation with LLMs for E-commerce

基于属性意识的LLM电商产品生成控制

Virginia Negri, Víctor Martínez Gómez, Sergio A. Balanya, Subburam Rajaram

机构 * Amazon Spain（亚马逊西班牙分公司） ； Amazon Germany（亚马逊德国分公司） 纠错

AI总结 本文提出基于LLM的属性意识电商产品生成方法，通过三种策略生成高质量合成数据，提升电商数据集的准确率与实用性。

Comments AAAI'26 Workshop on Shaping Responsible Synthetic Data in the Era of Foundation Models (RSD)

Beyond Detection: Exploring Evidence-based Multi-Agent Debate for Misinformation Intervention and Persuasion

超越检测：探索基于证据的多智能体辩论用于虚假信息干预与说服

Chen Han, Yijia Ma, Jin Tan, Wenzhen Zheng, Xijin Tang

AI总结 本文提出ED2D框架，通过整合事实证据检索，提升虚假信息检测与说服效果，同时揭示MAD系统在干预中的潜力与风险。

Comments This paper has been accepted to AAAI 2026

CaTFormer: Causal Temporal Transformer with Dynamic Contextual Fusion for Driving Intention Prediction

CaTFormer：具有动态上下文融合的因果时间变换器用于驾驶意图预测

Sirui Wang, Zhou Guan, Bingxi Zhao, Tongjia Gu, Jie Liu

AI总结 CaTFormer通过动态上下文融合和因果建模，提升驾驶意图预测的准确性和透明性。

Comments Accepted at AAAI 2026

PosterVerse: A Full-Workflow Framework for Commercial-Grade Poster Generation with HTML-Based Scalable Typography

PosterVerse: 一个用于商业级海报生成的全流程框架，基于HTML的可扩展字体

Junle Liu, Peirong Zhang, Yuyi Zhang, Pengyu Yan, Hui Zhou, Xinyue Zhou, Fengjun Guo, Lianwen Jin

AI总结 PosterVerse通过全流程框架和HTML驱动技术，实现商业级海报的高密度文本渲染与灵活设计。

Journal ref AAAI 2026 Oral

Evaluating LLMs for Police Decision-Making: A Framework Based on Police Action Scenarios

评估大语言模型用于警务决策：基于警务行动场景的框架

Sangyub Lee, Heedou Kim, Hyeoncheol Kim

AI总结 本文提出PAS框架，用于评估大语言模型在警务决策中的应用，揭示其在警务任务中的不足，并发布相关数据和模板以促进进一步研究。

Comments This work was accepted at AAAI 2026 social good track

Personalization of Large Foundation Models for Health Interventions

为健康干预个性化设计大型基础模型

Stefan Konigorski, Johannes E. Vedder, Babajide Alamu Owoyele, İbrahim Özkan

AI总结 本文提出混合框架，结合LFMs和N-of-1试验，以实现个性化医疗中的因果推断与责任AI整合。

Comments Accepted to the AAAI 2026 Workshop on Personalization in the Era of Large Foundation Models (PerFM)

EFX and PO Allocation Exists for Two Types of Goods

EFX和PO分配存在两种商品的情况

Vladimir Davidiuk, Yuriy Dementiev, Artur Ignatiev, Danil Sagunov

AI总结 本文证明在两种类型商品情况下，EFX和PO分配总能存在，并提出高效算法实现。

Comments Accepted at AAAI 2026

Exploration Through Introspection: A Self-Aware Reward Model

反思中的探索：一种自知奖励模型

Michael Petrowski, Milica Gašić

AI总结 本文提出一种基于反思的奖励模型，通过模拟疼痛信号探索自我意识对强化学习代理学习能力的影响，并展示了其在复杂行为复制中的优势。

Comments Accepted at AAAI-26 ToM4AI Workshop

Mass Concept Erasure in Diffusion Models with Concept Hierarchy

扩散模型中的概念擦除与概念层次

Jiahang Tu, Ye Li, Yiming Wu, Hanbin Zhao, Chao Zhang, Hui Qian

AI总结 本文提出了一种基于概念层次的扩散模型擦除方法，通过群体擦除和超类型保留低秩适应技术，提高擦除效率并减少生成质量退化。

Comments This paper has been accepted by AAAI 2026

Enhancing Noise Resilience in Face Clustering via Sparse Differential Transformer

通过稀疏微分变换器增强面部聚类的噪声鲁棒性

Dafeng Zhang, Yongqi Song, Shizhuo Liu

机构 * Dafeng Zhang, Yongqi Song, Shizhuo Liu（作者） 纠错

AI总结 本文提出稀疏微分变换器提升面部聚类的噪声鲁棒性，通过改进相似度测量方法以提高聚类性能。

Comments Published at AAAI 2026