arXivDaily arXiv每日学术速递 周一至周五更新

期刊&会议

AAAI Conference on Artificial Intelligence · 会议 · Artificial Intelligence

2026-01-21 至 2026-01-21 共收录 47
2601.11998 2026-01-21 cs.CR cs.AI

Hybrid IDS Using Signature-Based and Anomaly-Based Detection

基于签名检测和异常检测的混合入侵检测系统

Messaouda Boutassetta, Amina Makhlouf, Newfel Messaoudi, Abdelmadjid Benmachiche, Ines Boutabia

AI总结 本文提出一种结合签名检测和异常检测的混合IDS,旨在提升对新兴网络攻击的检测能力,并探讨其在金融、交通和社交网络中的应用。

Comments 7 pages,The Second National Conference on Artificial Intelligence and Information Technologies (NCAIIT25)

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.11903 2026-01-21 cs.AI

AEMA: Verifiable Evaluation Framework for Trustworthy and Controlled Agentic LLM Systems

AEMA:可验证评估框架用于可信和受控的代理LLM系统

YenTing Lee, Keerthi Koneru, Zahra Moslemi, Sheethal Kumar, Ramesh Radhakrishnan

机构 * University of California, San Diego(加州大学圣地亚哥分校) Center for Advanced AI, Accenture(Accenture高级人工智能中心) University of California, Irvine(加州大学伊拉斯姆斯分校)

AI总结 AEMA提出了一种可验证的评估框架,用于评估基于LLM的多代理系统,通过人类监督实现稳定、可追溯的自动化评估。

Comments Workshop on W51: How Can We Trust and Control Agentic AI? Toward Alignment, Robustness, and Verifiability in Autonomous LLM Agents at AAAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.11885 2026-01-21 cs.AI

MyGram: Modality-aware Graph Transformer with Global Distribution for Multi-modal Entity Alignment

MyGram: 多模态实体对齐的模态感知图变换器与全局分布

Zhifei Li, Ziyue Qin, Xiangyu Luo, Xiaoju Hou, Yue Zhao, Miao Zhang, Zhifang Huang, Kui Xiao, Bing Yang

AI总结 MyGram通过模态感知图变换器与全局分布机制,提升多模态实体对齐的性能。

Comments Accepted by AAAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.11872 2026-01-21 cs.CL

GloCTM: Cross-Lingual Topic Modeling via a Global Context Space

GloCTM:通过全局上下文空间进行跨语言主题建模

Nguyen Tien Phat, Ngo Vu Minh, Linh Van Ngo, Nguyen Thi Ngoc Diep, Thien Huu Nguyen

AI总结 GloCTM通过构建统一语义空间,实现跨语言主题建模的连贯性和语义对齐,提升多语言理解能力。

Comments AAAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.11865 2026-01-21 cs.CL

CTPD: Cross Tokenizer Preference Distillation

CTPD: 跨分词器偏好蒸馏

Truong Nguyen, Phi Van Dat, Ngan Nguyen, Linh Ngo Van, Trung Le, Thanh Hong Nguyen

AI总结 CTPD提出一种跨分词器的偏好蒸馏框架,通过对齐跨度投影、跨分词器重要性采样和教师锚定参考,实现不同分词器模型间的偏好信息高效转移。

Comments AAAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.11821 2026-01-21 cs.LG

Shapelets-Enriched Selective Forecasting using Time Series Foundation Models

基于时间序列基础模型的形状let增强选择性预测

Shivani Tomar, Seshu Tirupathi, Elizabeth Daly, Ivana Dusparic

AI总结 本文提出基于形状let的选性预测框架,利用时间序列基础模型减少预测误差,提升模型可靠性。

Comments Accepted by the AAAI-26 Workshop on Artificial Intelligence for Time Series Analysis (AI4TS)

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.11816 2026-01-21 cs.AI

POLARIS: Typed Planning and Governed Execution for Agentic AI in Back-Office Automation

POLARIS:面向后台自动化代理AI的类型化规划与受控执行

Zahra Moslemi, Keerthi Koneru, Yen-Ting Lee, Sheethal Kumar, Ramesh Radhakrishnan

机构 * University of California, Irvine(加州大学尔湾分校) Center for Advanced AI, Accenture(Accenture高级人工智能中心) University of California, San Diego(加州大学圣地亚哥分校)

AI总结 POLARIS通过类型化规划和受控执行框架,提升后台自动化中代理AI的政策一致性与可预测性,实现高精度任务处理与审计追踪。

Comments Workshop on Agentic AI Benchmarks and Applications for Enterprise Tasks: AAAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.11567 2026-01-21 cs.CL cs.AI

Measuring Stability Beyond Accuracy in Small Open-Source Medical Large Language Models for Pediatric Endocrinology

在儿科内分泌学中超越准确性的医疗小型开源大语言模型稳定性测量

Vanessa D'Amario, Randy Daniel, Alessandro Zanetti, Dhruv Edamadaka, Nitya Alaparthy, Joshua Tarkoff

AI总结 本文研究了儿科内分泌学中医疗小型开源大语言模型的稳定性,发现高一致性不等于正确性,且系统扰动会影响输出,强调了评估框架的必要性。

Comments 20 pages, 11 figures, accepted at 47 workshop Reproducible Artificial Intelligence (AAAI 2026, Singapore, January 27, 2026)

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.11545 2026-01-21 cs.HC

Multimodal Data Fusion to Capture Dynamic Interactions between Built Environment and Vulnerable Older Adults

多模态数据融合以捕捉建成环境与易受伤害的老年人之间的动态交互

Houhao Liang, Azrin Jamaluddin, Kresimir Friganovic, Kirstie Neo, Raphael Han, Navrag Singh, Panos Mavros

AI总结 本研究通过多模态数据融合技术,探索建成环境对易受伤害老年人移动性的影响,为包容性城市规划提供数据支持。

Comments This work has been accepted to the AAAI 2026 Workshop on AI for Urban Planning

详情

展开后加载摘要…

URL PDF HTML 收藏
2601.11526 2026-01-21 cs.HC cs.AI

Chatsparent: An Interactive System for Detecting and Mitigating Cognitive Fatigue in LLMs

Chatsparent: 一个用于检测和缓解大语言模型认知疲劳的交互系统

Riju Marwah, Vishal Pallagani, Ritvik Garimella, Amit Sheth

AI总结 Chatsparent通过实时监测和缓解大语言模型的认知疲劳,提升交互体验和模型可靠性。

Comments Accepted to AAAI 2026 Demonstration Track

详情

展开后加载摘要…

URL PDF HTML 收藏
2512.03994 2026-01-21 cs.LG

Training-Free Policy Violation Detection via Activation-Space Whitening in LLMs

无需训练的策略违规检测:通过激活空间白化在大语言模型中

Oren Rachmil, Avishag Shapira, Roy Betser, Itay Gershon, Omer Hofman, Asaf Shabtai, Yuval Elovici, Roman Vainshtein

机构 * Fujitsu Research of Europe(富士通欧洲研究机构) Ben-Gurion University of the Negev(贝内杰尔大学)

AI总结 本文提出一种无需训练的策略违规检测方法,通过激活空间白化技术在大语言模型中实现高效检测。

Comments Accepted to the AAAI 2026 Deployable AI (DAI) Workshop

详情

展开后加载摘要…

URL PDF HTML 收藏
2510.00332 2026-01-21 cs.AI cs.CE

When Hallucination Costs Millions: Benchmarking AI Agents in High-Stakes Adversarial Financial Markets

当幻觉成本百万:在高风险对抗性金融市场中基准测试AI代理

Zeshi Dai, Zimo Peng, Zerui Cheng, Ryan Yihe Li

机构 * Surf AI, Cybertino Lab(Surf AI,Cybertino 实验室) Princeton University(普林斯顿大学)

AI总结 CAIA基准测试揭示了AI在对抗性金融市场中的能力缺口,指出当前模型在面对虚假信息和不可逆决策时表现不佳,强调对抗鲁棒性对可信AI的重要性。

Comments 15 pages, 5 figures, 4 tables; Accepted to AAAI 2026 (AI-4-Finance Workshop - Oral, top 10%); In submission to ICML 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.18260 2026-01-21 cs.CL

MIRAGE: Scaling Test-Time Inference with Parallel Graph-Retrieval-Augmented Reasoning Chains

MIRAGE: 通过并行图检索增强推理链实现测试时扩展

Kaiwen Wei, Rui Shan, Dongsheng Zou, Jianzhong Yang, Bi Zhao, Junnan Zhu, Jiang Zhong

AI总结 MIRAGE通过并行图检索增强推理链实现测试时扩展,提升医疗问答任务的准确性和可追溯性。

Comments 10 pages, 8 figures (including tables), plus appendix. Accepted to AAAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.13866 2026-01-21 cs.CV

SAGA: Learning Signal-Aligned Distributions for Improved Text-to-Image Generation

SAGA:学习信号对齐的分布以提高文本到图像生成

Paul Grimal, Michaël Soumm, Hervé Le Borgne, Olivier Ferret, Akihiro Sugimoto

机构 * Université Paris-Saclay, CEA, List(巴黎-萨克雷大学,CEA,List) Télécom Paris(巴黎电信学院) National Institute of Informatics, Japan(日本信息处理研究所)

AI总结 SAGA 通过学习信号对齐的分布,提高文本到图像生成的准确性与控制性,支持多种条件模式并优于现有方法。

Comments Accepted to AAAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.04444 2026-01-21 cs.LG cs.NA math.NA stat.ML

Matrix-Free Two-to-Infinity and One-to-Two Norms Estimation

无矩阵的二到无穷范数和一到二范数估计

Askar Tsyganov, Evgeny Frolov, Sergey Samsonov, Maxim Rakhuba

AI总结 本文提出无矩阵设置下的二到无穷范数和一到二范数估计新算法,基于Hutchinson估计器的改进,并展示其在深度学习和推荐系统中的应用。

Comments AAAI-2026, camera-ready version

详情

展开后加载摘要…

URL PDF HTML 收藏
2508.01412 2026-01-21 cs.CL

Bias Association Discovery Framework for Open-Ended LLM Generations

面向开放型大语言模型生成的偏见关联发现框架

Jinhao Pan, Chahat Raj, Ziwei Zhu

机构 * Jinhao Pan Chahat Raj Ziwei Zhu

AI总结 本研究提出BADF框架,用于从开放型LLM生成中发现和分析人口身份与描述性概念的偏见关联。

Comments AAAI 2026

详情

展开后加载摘要…

URL PDF HTML 收藏
2505.23816 2026-01-21 cs.CL cs.LG

A Course Correction in Steerability Evaluation: Revealing Miscalibration and Side Effects in LLMs

在可转向性评估中的课程修正:揭示LLMs中的误校准与副作用

Trenton Chang, Tobias Schnabel, Adith Swaminathan, Jenna Wiens

AI总结 本文提出了一种多维目标空间框架,揭示LLMs在文本改写任务中存在意外副作用,表明现有对齐策略可能不足。

Comments 8 pages, 6 figures. 26 pages of references and supplementary material, 22 additional figures. Association for the Advancement of Artificial Intelligence Conference (AAAI 2026)

详情

展开后加载摘要…

URL PDF HTML 收藏