多模态大模型

Inconsistency-aware Multimodal Schrödinger Bridge for Deepfake Localization

不一致感知多模态薛定谔桥用于深度伪造定位

Jiayu Xiong, Jing Wang, Qi Zhang, Wanlong Wang, Jun Xue

机构 * Department of Computer Science and Techonology, Huaqiao University（华侨大学计算机科学与技术系） ； Xiamen Key Laboratory of Computer Vision and Pattern Recognition, Huaqiao University（厦门计算机视觉与模式识别重点实验室） ； Tongji University（同济大学） ； School of Cyber Science and Engineering, Wuhan University（武汉大学网络空间安全学院） 纠错

专题命中 音频语音多模态 ：multimodal(title,abstract)；cross-modal(abstract)；audio-visual(abstract)；分类 cs.CV

AI总结 提出IaMSB框架，利用薛定谔桥统一一致性估计、跨模态信息选择和桥步调度，实现高精度区间级深度伪造定位。

Comments Accepted by CVPR2026

Benchmarking Gaslighting Attacks Against Speech Large Language Models

针对语音大语言模型的气灯攻击基准测试

Jinyang Wu, Bin Zhu, Xiandong Zou, Qiquan Zhang, Xu Fang, Pan Zhou

机构 * Singapore Management University（新加坡管理学院） ； Tongyi Speech Lab（通义语音实验室） ； Dalian University of Technology（大连理工大学） 纠错

专题命中 音频语音多模态 ：multi-modal(abstract)；分类 cs.CL

AI总结 提出五种气灯攻击策略（愤怒、认知干扰、讽刺、隐晦、专业否定），在5个语音和多模态大语言模型上测试，发现平均准确率下降24.3%，揭示语音模型的行为脆弱性。

Comments 5 pages, 2 figures, 3 tables

StepAudio 2.5 Technical Report

StepAudio 2.5 技术报告

Bin Lin, Bo Zhao, Boyong Wu, Chao Yan, Chen Wu, Cheng Yi, Chengyuan Yao, Daijiao Liu, Fei Tian, Feng Tian, Haiyang Sun, Haoyang Zhang, Jiangjie Zhen, Jinglan Gong, Jun Chen, Li Xie, Peilin Li, Peng Yang, Pengfei Tan, Qingjian Lin, Runze Li, Shenghua Hu, Siyi Zhou, Wenwen Qu, Xiangyu Li, Xiangyu Tony Zhang, Xuerui Yang, Yang Yang, Yechang Huang, Yu Fu, Yuchu Luo, Yuxin Li, Yuxin Zhang, Zhengyan Sheng, Brian Li, Chang Zeng, Changlin Zhang, Chen Geng, Chenghao Dong, Chengli Feng, Dan Zhou, Danni Wan, Di Chen, Die Zhang, Dongqing Pang, Guanglong Yang, Guoqiang Hu, Huangxi Zhu, Jianzheng Gao, Jinghua Liang, Jinmei Wan, Junjie Yuan, Kang An, Lei Lei, Limin Zhong, Lun Cai, Mengqiang Ren, Min Xu, Mingliang Li, Mingxiao Li, Na Wang, Qiang Tong, Qiaoling Huang, Qingfu Du, Rui Wang, Shengchen Zhou, Shi Qiu, Shihao Peng, Shiliang Yang, Siqi Tu, Tianjiao Deng, Ting Xu, Tong Wang, WeiMing Niu, Wuxun Xie, Xianwei Zhang, Xianyu Feng, Xiaojia Liu, Xing Chen, Xiongbin Wu, Yan Wu, Yang Li, Yi Liu, Yifan Zhang, Yile Liu, Yongshen Long, Yu Luo, Yuanhao Ding, Yuhao Wang, Yuhe Yin, Yunfang Xu, Yuxiang Yang, Zhiguo Huang, Zhiyue Wu, Zichao Li, Zichao Zhou, Daxin Jiang, Future Li, Gang Yu, Xiangyu Zhang, Yibo Zhu

专题命中 音频语音多模态 ：multimodal(abstract)；分类 eess.AS

AI总结 提出统一音频语言基础模型StepAudio 2.5，通过任务定制的RLHF后训练和专用解码策略，在ASR、TTS和实时口语交互三项任务上达到或超越专用系统水平。