多模态大模型

Improving Large Vision-Language Models' Understanding for Flow Field Data

提升大型视觉-语言模型对流场数据的理解能力

Xiaomei Zhang, Hanyu Zheng, Xiangyu Zhu, Jinghuan Wei, Junhong Zou, Zhen Lei, Zhaoxiang Zhang

机构 * State Key Laboratory of Multimodal Artificial Intelligence Systems, Institute of Automation, Chinese Academy of Sciences（多模态人工智能系统国家重点实验室，自动化研究所，中国科学院） 纠错

专题命中 图文多模态 ：multimodal(abstract)；分类 cs.CV

AI总结 FieldLVLM通过场感知语言生成策略和数据压缩多模态模型调优，提升大型视觉-语言模型对流场数据的理解能力。

Comments Accepted by Machine Intelligence Research

CLIP: Lightweight Cosine-Law-Based Inverted-List Pruning for IVF-Based Vector Search

CLIP：基于余弦定律的轻量级倒排列表剪枝技术用于IVF向量搜索

Yitong Song, Shuhang Lu, Pengcheng Zhang, Jianliang Xu

专题命中 图文多模态 ：multimodal(abstract)

AI总结 针对IVF向量搜索中粗粒度执行导致高延迟的问题，提出基于余弦定律的轻量级剪枝技术CLIP，支持簇间和簇内剪枝，显著减少不必要的簇和向量访问，并开发了IVF-CLIP、HIVF-CLIP和LSM-IVF三种变体，实验表明剪枝率达78%，效率提升最高141%。

SocialPulse: On-Device Detection of Social Interactions in Naturalistic Settings Using Smartwatch Sensing

SocialPulse: 使用智能手表多模态感知在自然环境中检测社交互动

Md Sabbir Ahmed, Kaitlyn Dorothy Petz, Noah French, Tanvi Lakhtakia, Aayushi Sangani, Mark Rucker, Xinyu Chen, Bethany A. Teachman, Laura E. Barnes

专题命中 音频语音多模态 ：multimodal(title)

AI总结 本文提出了一种在设备上检测自然环境中社交互动的方法，通过智能手表的多模态感知技术，实现了对多种社交互动的准确识别，并在真实场景中验证了其有效性。

Gemma 4 Technical Report

Gemma 4技术报告

Gemma Team, Sherif El Abd, Vaibhav Aggarwal, Robin Algayres, Alek Andreev, Olivier Bachem, Ian Ballantyne, Cormac Brick, Victor Cărbune, Michelle Casbon, Mayank Chaturvedi, Aditya Chawla, Victor Cotruta, Alice Coucke, Phil Culliton, Robert Dadashi, Lucas Dixon, Mohamed Elhawaty, Utku Evci, Clément Farabet, Johan Ferret, Filippo Galgani, Sertan Girgin, Jean-Bastien Grill, Maarten Grootendorst, Jiaxian Guo, Cassidy Hardin, Yanzhang He, Steven M. Hernandez, Omri Homburger, Léonard Hussenot, Juyeong Ji, Armand Joulin, Aishwarya Kamath, Parnian Kassraie, Olivier Lacombe, Preethi Lahoti, Gaël Liu, Gus Martins, Luciano Martins, Tatiana Matejovicova, Ramona Merhej, Nikola Momchev, Sneha Mondal, Ryan Mullins, Sindhu Raghuram Panyam, Shreya Pathak, Sarah Perrin, André Susano Pinto, Etienne Pot, Angéline Pouget, Alexandre Ramé, Sabela Ramos, Douglas Reid, David Rim, Morgane Rivière, Karsten Roth, Louis Rouillard, Omar Sanseviero, Pier Giuseppe Sessa, Shane Settle, Danila Sinopalnikov, Sara Smoot, Piotr Stanczyk, Andreas Steiner, Lawrence Stewart, Ilya Tolstikhin, Michael Tschannen, Anton Tsitsulin, Nino Vieillard, Renjie Wu, Pingmei Xu, Haichuan Yang, Edouard Yvinec, Biao Zhang, Li Zhang, Joe Zou, Nicolas Aagnes, Abdelrahman Abdelhamed, Jakub Adamek, Shivani Agrawal, Shubham Agrawal, Ibrahim Alabdulmohsin, Jean Baptiste Alayrac, Uri Alon, Chandramouli Amarnath, Ankesh Anand, Chrysovalantis Anastasiou, Setareh Ariafar, François-Xavier Aubet, Kyriakos Axiotis, Federico Barbero, Joelle Barral, Alexei Bendebury, Urs Bergmann, Stanley Bileschi, Kat Black, Mathieu Blondel, Sebastian Borgeaud, Arthur Bražinskas, Ryan Burnell, Robert Busa-Fekete, Mu Cai, Daniele Calandriello, Glenn Cameron, Charlotte Caucheteux, Rahma Chaabouni, Garima Chadha, Jetha Chan, Blake Jianhang Chen, Jesse Chen, Lin Chen, Xu Chen, Derek Cheng, Tzu-hsiang Chien, Nikolai Chinaev, Yi Chou, Zhaohui Chu, Benjamin Coleman, Pooja Consul, Sam Conway-Rahman, Scott Crowell, Dylan Cutler, Vivek Dani, Samira Daruki, Anil Das, Daniel Deutsch, Nishanth Dikkala, Li Ding, Qiuhan Ding, Shenil Dodhia, Konstantin Donhauser, Tulsee Doshi, Anca Dragan, Alex Druinsky, Sahil Dua, Zoltan Egyed, Danielle Eisenbud, Daniel Eppens, Cindy Fan, Bahare Fatemi, Yassir Fathullah, Vlad Feinberg, Milen Ferev, Sebastian Flennerhag, Takumi Fujimoto, João Gabriel Oliveira, Isaac Galatzer-Levy, João Gante, Simon Geisler, Soham Ghosal, Antonious M. Girgis, Tamara von Glehn, Alec Go, Alhaad Gokhale, Alex Grills, Yiming Gu, Mayank Gupta, Pramod Gupta, Guru Guruganesh, Raia Hadsell, Hamza Harkous, Jitendra Harlalka, Demis Hassabis, Anja Hauth, Joe Heyward, Arian Hosseini, Chih-Yang Hsia, I-Hung Hsu, Xiaopeng Huang, Yangsibo Huang, Kevin Hui, Adrian Hutter, Te I, Fotis Iliopoulos, Advait Jain, Ganesh Jawahar, Ziwei Ji, Qilin Jin, Melvin Johnson, Kandarp Joshi, Arun Kandoor, Wang-Cheng Kang, Koray Kavukcuoglu, Mehran Kazemi, Kathleen Kenealy, Amr Khalifa, Phoebe Kirk, Ivan Korotkov, Suraj Kothawade, Vitaly Kovalev, Neel Kovelamudi, Adam Kraft, Ravin Kumar, Vivek Kumar, Harish Kuppam, Justin Lannin, Chen-Yu Lee, Seungji Lee, Dmitry Lepikhin, Alon Levkovitch, Dongdong Li, Qiujia Li, Valentin Liévin, Ethan Lin, Ziqian Lin, Casper Liu, Tianlin Liu, Tianqi Liu, Xin Liu, Ivan Lobov, Mayank Lunayach, Min Ma, Gagan Madan, Andrii Maksai, Eric Malmi, Michal Matuszak, Daniel McDuff, Gaurav Menghani, Maciej Mikuła, Daniil Mirylenka, Karolis Misiunas, Vedant Misra, Andreea Mitran, Kareem Mohamed, Maksim Mukha, Eric Noland, James O'Donnell, Brendan O'Donoghue, Kate Olszewska, Bernett Orlando, Wanqiong Pan, Rina Panigrahy, Unnati Parekh, Nicolas Perez-Nieves, Chunjong Park, Eric Paskie, Liqian Peng, Bryce Petrini, Slav Petrov, Jonas Pfeiffer, Bilal Piot, Martyna Plomecka, Siim Poder, Octavio Ponce, Arijit Pramanik, David Racz, Anish Rajan, Michelle Ramanovich, Anand Rao, Marvin Ritter, Vitor Rodrigues, Evan Rosen, Mikołaj Rybiński, Noveen Sachdeva, Michaël E. Sander, Rohit Sathyanarayana, Sagar Savla, Samuel Schmidgall, Tal Schuster, George Scrivener, Benoit Seguin, Andrew Sellergren, Aliaksei Severyn, Izhak Shafran, Dhruv Shah, Bobak Shahriari, Yuan Shangguan, Ashish Shenoy, Pradeep Shenoy, Rakesh Shivanna, Pauline Sho, Lucas Spangher, Wojciech Stokowiec, Tim Strother, Yao Su, Yinghao Sun, Mukund Sundararajan, Andrea Tacchetti, Mor Hazan Taege, Pouya Tafti, Jean Tarbouriech, Chetan Tekur, Shantanu Thakoor, Rahul Thapa, Madeleine Traverse, Lenart Treven, Tao Tu, Chien Te Tung, Çağlar Ünlü, Petar Veličković, Malini Pooni Venkat, Sagar Gubbi Venkatesh, Vidya Venkiteswaran, Francesco Visin, Alex Vitvitskyi, Kiran Vodrahalli, Weiyi Wang, Xin Wang, Tris Warkentin, Jan Wassenberg, John Wieting, Cindy Wu, Lechao Xiao, Hao Xu, Yuhui Xu, Fuzhao Xue, Arun Yadav, Jun Yan, Antoine Yang, Lin Yang, Ming-Hsuan Yang, Ziyu Ying, Jae Hyeon Yoo, Morteza Zadimoghaddam, Sajjad Zafar, Fred Zhang, Jiageng Zhang, Jianyi Zhang, Xiaofan Zhang, Chao Zhao, David Zhou, Chen Zou

机构 * Google DeepMind（谷歌DeepMind） 纠错

专题命中 音频语音多模态 ：multimodal(abstract)；分类 cs.CL、cs.AI

AI总结 介绍新一代Gemma 4开源多模态语言模型，通过密集和专家混合架构提升计算效率与推理能力，提出统一无编码器架构，集成思考模式，改进多方面性能，在多基准测试中有显著提升。

Comments 17 pages, 2 figures, technical report, updated

Local Multimodal Music Alignment from Global Supervision

基于全局监督的局部多模态音乐对齐

Irmak Bukey, Zachary Novack, Jongmin Jung, Dasaem Jeong, Chris Donahue

专题命中 跨模态检索 ：multimodal(title,abstract)；cross-modal(abstract)；分类 cs.MM

AI总结 研究如何利用全局监督实现局部多模态音乐对齐，提出FuSiLi方法，通过基于Sinkhorn的软对齐操作局部特征，结合传统全局相似性微调预训练编码器，在跨模态检索和帧级对齐任务中表现出色，优于多种基线。

Comments ISMIR 2026

HunyuanVideo-HOMA: Generic Human-Object Interaction in Multimodal Driven Human Animation

混元视频-HOMA：多模态驱动人体动画中的通用人机交互

Ziyao Huang, Zixiang Zhou, Juan Cao, Yifeng Ma, Yi Chen, Zejing Rao, Zhiyong Xu, Hongmei Wang, Qin Lin, Yuan Zhou, Qinglin Lu, Fan Tang

机构 * University of Chinese Academy of Sciences（中国科学院大学） ； Tencent Hunyuan（腾讯混元） 纠错

专题命中 多模态生成 ：multimodal(title,abstract)；分类 cs.CV

AI总结 针对人体-物体交互视频生成的局限，提出混元视频-HOMA弱条件多模态驱动框架，通过稀疏解耦运动引导等方法，将外观和运动信号编码融合，经优化训练，在弱监督下性能达先进水平，还具文本生成和物体操纵通用性。

Now We Know? A Systematic Comparison of TerraMind and THOR

我们现在知道了吗？TerraMind和THOR的系统比较

Frederick Schindlegger, Kenzo Bounegta, Eva Gmelich Meijling, Johannes Jakubik, Arnt-Børre Salberg, Theodor Forgaard, Nicolas Longepe, Valerio Marsocci

机构 * University of Münster（明斯特大学） ； IBM Research（IBM研究院） ； Norwegian Computing Center（挪威计算中心） 纠错

专题命中 多模态生成 ：multimodal(abstract)；cross-modal(abstract)；any-to-any(abstract)；分类 cs.CV、cs.AI

AI总结 通过对TerraMind和THOR两个地理空间基础模型对比，研究其在补丁大小、解码器复杂性等方面的差异轴，发现架构设计选择对性能差异影响更大，体现互补投资策略，还得出假设和诊断消融方法，有望推广到未来模型。

GeoDiff-SAR: A Geometric Prior Guided Diffusion Model for SAR Image Generation

GeoDiff-SAR：一种基于几何先验的扩散模型用于SAR图像生成

Fan Zhang, Xuanting Wu, Fei Ma, Qiang Yin, Yuxin Hu

机构 * College of Information Science and Technology, Beijing University of Chemical Technology（信息科学与技术学院，北京化工大学） ； Aerospace Information Research Institute, Chinese Academy of Sciences（航天信息研究所，中国科学院） 纠错

专题命中 多模态生成 ：multi-modal(abstract)；分类 cs.CV

AI总结 GeoDiff-SAR通过引入几何先验引导扩散模型，提升SAR图像生成的保真度和分类准确性，尤其在不同方位角识别性能上有显著提升。

Comments 3 pages, 1 figures

PixDLM: A Dual-Path Multimodal Language Model for UAV Reasoning Segmentation

PixDLM：一种用于无人机推理分割的双路径多模态语言模型

Shuyan Ke, Yifan Mei, Changli Wu, Yonghan Zheng, Jiayi Ji, Liujuan Cao, Rongrong Ji

机构 * Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University（中国教育部多媒体可信感知与高效计算重点实验室，厦门大学） 纠错

专题命中 多模态评测 ：multimodal(title,abstract)；分类 cs.CV

AI总结 本文提出PixDLM，一种用于无人机推理分割的多模态语言模型，通过构建DRSeg基准数据集，验证了该模型在处理高分辨率无人机图像中的有效性。

Comments Accepted to CVPR 2026 (highlight)

The Computational Basis of Confidence in Large Language Models

大语言模型中置信度的计算基础

Dharshan Kumaran, Viorica Patraucean, Maks Ovsjanikov, Petar Veličković, Nathaniel Daw

机构 * Google DeepMind（谷歌DeepMind） ； École Polytechnique（巴黎综合理工学院） ； Princeton University（普林斯顿大学） 纠错

专题命中 多模态评测 ：multimodal(abstract)；分类 cs.AI

AI总结 研究大语言模型中置信度的计算基础，利用统计决策置信度框架，通过答案 - 对数差异测试其预测特征，结果表明在多种任务中答案对数可作潜在决策变量读出，为多模态语言模型置信度提供解释并统一研究框架。

Industrial Dexterity Benchmark: A Hardware-Software Benchmarking Platform for Industrial Dexterous Manipulation

工业灵巧性基准测试：一个用于工业灵巧操作的硬件-软件基准测试平台

Honglu He, Jacob Laufer, Zhiwu Zheng, David Elkan-gonzalez, Raman Goyal, Xinyi Li, Su Lu, Mishek Musa, Berke Saat, Nicolas Tan, Colm Prendergast

机构 * Analog Devices, Inc.（亚德诺半导体技术有限公司） 纠错

专题命中 多模态评测 ：multimodal(abstract)

AI总结 针对工业灵巧操作瓶颈，提出从经典流程到端到端多模态模仿学习框架的转变，介绍了IDB板、DAG-ROS框架和AG-iDP3策略框架，通过数据中心电缆操作实验表明新策略在多方面优于传统方法，推动向可扩展机器人自动化发展。

Layer-wise LoRA fine-tuning: a similarity metric approach

逐层LoRA微调：相似性度量方法

Keith Ando Ogawa, Bruno Lopes Yamamoto, Lucas Lauton de Alcantara, Lucas Pellicer, Rosimeire Pereira Costa, Edson Bollis, Anna Helena Reali Costa, Artur Jordao

专题命中 多模态评测 ：multimodal(abstract)

AI总结 本文提出逐层LoRA微调方法，通过选择性微调部分层以减少可训练参数并保持预测性能，适用于不同架构和任务。

Comments Accepted at the International Conference on Pattern Recognition (ICPR 2026)

Moving Beyond Diversity: Visual Token Pruning as Subspace Reconstruction for Efficient VLMs

超越多样性：将视觉令牌剪枝视为子空间重建以实现高效视觉语言模型

Jaeyeon Lee, Shunjie Wen, Dong-Wan Choi

机构 * Inha University（延世大学） 纠错

专题命中 多模态训练与对齐 ：image-text(abstract)；分类 cs.CV

AI总结 提出SPARE方法，将令牌剪枝重构为子空间重建问题，通过迭代选择投影残差大的令牌进行剪枝，并引入反相关性机制保留上下文信息，在LLaVA上剪枝94%令牌仍保持95%性能。

Adaptive Learned State Estimation based on KalmanNet

基于KalmanNet的自适应学习状态估计

Arian Mehrfard, Bharanidhar Duraisamy, Stefan Haag, Florian Geiss, Mirko Mählisch

机构 * Mercedes-Benz AG（梅赛德斯-奔驰集团） ； Institute for Autonomous Driving, University of the Bundeswehr Munich（慕尼黑联邦国防军大学自动驾驶研究所） 纠错

专题命中 其他多模态 ：multi-modal(abstract)

AI总结 本文提出AM-KNet，通过引入传感器特定的测量模块和超网络实现多传感器自动驾驶中的自适应状态估计，提升真实世界汽车数据的估计精度和跟踪稳定性。

Comments Pending release approval from company