Ensembling Multiple Hallucination Detectors Trained on VLLM Internal Representations
集成多个基于VLLM内部表示的幻觉检测器
AI总结 本文提出通过集成多个基于VLLM内部表示的幻觉检测模型，以减少幻觉并提高VQA任务的准确性。
Comments 5th place solution at Meta KDD Cup 2025
期刊&会议
KDD
ACM SIGKDD Conference on Knowledge Discovery and Data Mining · 会议 · Data Mining
RIPCN: A Road Impedance Principal Component Network for Probabilistic Traffic Flow Forecasting
RIPCN: 一条道路阻抗主成分网络用于概率交通流预测
机构 * School of Computer Science ； Technology Beijing Jiaotong University Beijing China ； Department of Computer Science Aalborg University Aalborg Denmark ； Key Laboratory of Big Data \& Artificial Intelligence in Transportation, Ministry of Education Beijing China ； Beijing Key Laboratory of Traffic Data Mining ； Beijing Jiaotong University ； Aalborg University ； Key Laboratory of Big Data \& Artificial Intelligence in Transportation, Ministry of Education
AI总结 RIPCN通过结合交通理论与时空主成分学习，提升交通流预测的准确性和不确定性估计能力。
Comments Accepted at KDD 2026. 12 pages, 10 figures
TAMEing Long Contexts in Personalization: Towards Training-Free and State-Aware MLLM Personalized Assistant
在个性化中延长上下文：迈向无训练和状态感知的MLLM个性化助手
机构 * University of Electronic Science and Technology of China（电子科技大学） ； Aiwen Technology Co., Ltd.（Aiwen科技有限公司） ； Intelligent Digital Media Technology Key Laboratory of Sichuan Province（四川省智能数字媒体技术重点实验室）
AI总结 本文提出TAME框架，通过双记忆和RA2G范式实现无训练、状态感知的MLLM个性化，提升长上下文对话能力。
Comments Accepted by KDD 2026 research track. Code and data are available at https://github.com/ronpay/TAME
From Shallow Humor to Metaphor: Towards Label-Free Harmful Meme Detection via LMM Agent Self-Improvement
从浅层幽默到隐喻：通过LMM代理自我改进实现无标签有害迷因检测
机构 * University of Electronic Science and Technology of China（电子科技大学） ； Southwestern University of Finance and Economics（西南财经大学） ； Intelligent Digital Media Technology Key Laboratory of Sichuan Province（四川省智能数字媒体技术重点实验室）
AI总结 ALARM通过LMM代理自我改进实现无标签有害迷因检测，利用浅层迷因信息提升对复杂迷因的识别能力。
Comments 12 pages. Accepted by KDD 2026 research track. Codes are released at https://github.com/Jian-Lang/ALARM
