Recent advances in AI-generated content (AIGC) have led to the emergence of powerful text-to-video generation models. Despite these successes, evaluating the quality of AIGC-generated videos remains challenging due to limited generalization, lack of temporal awareness, heavy reliance on large-scale annotated datasets, and the lack of effective interaction with generation models. Most current approaches rely on supervised finetuning of vision-language models (VLMs), which often require large-scale annotated datasets and tend to decouple understanding and generation. To address these shortcomings, we propose VQ-Insight, a novel reasoning-style VLM framework for AIGC video quality assessment. Our approach features: (1) a progressive video quality learning scheme that combines image quality warm-up, general task-specific temporal learning, and joint optimization with the video generation model; (2) the design of multi-dimension scoring rewards, preference comparison rewards, and temporal modeling rewards to enhance both generalization and specialization in video quality evaluation. Extensive experiments demonstrate that VQ-Insight consistently outperforms state-of-the-art baselines in preference comparison, multi-dimension scoring, and natural video scoring, bringing significant improvements for video generation tasks.
Teaching VLMs to Localize Specific Objects from In-context Examples
IBM Research（IBM研究院）；Weizmann Institute of Science（魏茨曼科学研究所）；Tel Aviv University（特拉维夫大学）；JKU Linz（林茨约翰内斯·开普勒大学）；Tuebingen AI Center（蒂宾根人工智能中心）；MIT-IBM（麻省理工学院-IBM联合机构）；MIT CSAIL（麻省理工学院计算机科学与人工智能实验室）
Teaching Vision-Language Models to Ask: Resolving Ambiguity in Visual Questions
Institute of Automation, Chinese Academy of Sciences（中国科学院自动化研究所）；School of Artificial Intelligence, University of Chinese Academy of Sciences（中国科学院大学人工智能学院）；Wuhan AI Research（武汉人工智能研究院）
Google Research（谷歌研究院）；Tel Aviv University（特拉维夫大学）；Weizmann Institute of Science（魏茨曼科学研究所）
Large vision-language models (VLMs), such as CLIP, learn rich joint image-text representations, facilitating advances in numerous downstream tasks, including zero-shot classification and text-to-image generation. Nevertheless, existing VLMs exhibit a prominent well-documented limitation - they fail to encapsulate compositional concepts such as counting. We introduce a simple yet effective method to improve the quantitative understanding of VLMs, while maintaining their overall performance on common benchmarks. Specifically, we propose a new counting-contrastive loss used to finetune a pre-trai…
PhyBlock: A Progressive Benchmark for Physical Understanding and Planning via 3D Block Assembly
Mohamed bin Zayed University of Artificial Intelligence（莫扎德·本·扎耶德人工智能大学）；Tsinghua Shenzhen International Graduate School, Tsinghua University, China（清华大学深圳国际研究生院）；Shenzhen Campus of Sun Yat-sen University（孙中山大学深圳校区）；Spatialtemporal AI（时空人工智能）；Shanghai Jiao Tong University（上海交通大学）
SECOND: Mitigating Perceptual Hallucination in Vision-Language Models via Selective and Contrastive Decoding
Department of Electrical and Computer Engineering, Seoul National University（首尔大学电气与计算机工程系）；Interdisciplinary Program in Artificial Intelligence, Seoul National University（首尔大学人工智能跨学科项目）
ProAPO: Progressively Automatic Prompt Optimization for Visual Classification
Institute of Information Engineering, Chinese Academy of Sciences（中国科学院信息工程研究所）；School of Cyber Security, University of Chinese Academy of Sciences（中国科学院大学网络安全学院）；School of Information Engineering, Minzu University of China（民族大学信息工程学院）；University of Science and Technology Beijing（北京科技大学）
Advancing Complex Wide-Area Scene Understanding with Hierarchical Coresets Selection
Institute of Software Chinese Academy of Sciences（中国科学院软件研究所）；University of the Chinese Academy of Sciences（中国科学院大学）；Beijing University of Technology（北京理工大学）
MovieTeller: Tool-augmented Movie Synopsis with ID Consistent Progressive Abstraction
Central Universities（中央高校）；National Natural Science Foundation of China（中国国家自然科学基金委员会）；Research Fund for International Scientists of National Natural Science Foundation of China（中国国家自然科学基金委员会国际科学家研究基金）
Astrea: A MOE-based Visual Understanding Model with Progressive Alignment
Zhejiang University（浙江大学）；Southern University of Science and Technology（南方科技大学）；Beijing University of Technology（北京工业大学）；Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Hong Kong Polytechnic University（香港理工大学）；Qingdao University（青岛大学）
What Happens Before Decoding? Prefill Determines GUI Grounding in VLMs
Guangming Laboratory（光明实验室）；National University of Singapore（新加坡国立大学）；Peking University（北京大学）；University of Waterloo（滑铁卢大学）；The Chinese University of Hong Kong (Shenzhen)（香港中文大学（深圳））
Synthesize, Diagnose, and Optimize: Towards Fine-Grained Vision-Language Understanding
Fudan University（复旦大学）；NVIDIA（英伟达）；Shanghai Key Lab of Intell. Info. Processing（上海智能信息处理重点实验室）；Shanghai Collaborative Innovation Center of Intelligent Visual Computing（上海智能视觉计算协同创新中心）
P-MTP: Efficient Document Parsing via Multi-Token Prediction with Progressive Depth Scaling
Department of Computer Vision Technology (VIS) Baidu Inc China（百度计算机视觉技术部（VIS））；Tsinghua University Shenzhen International Graduate School China（清华大学深圳国际研究生院）
PVC: Progressive Visual Token Compression for Unified Image and Video Processing in Large Vision-Language Models
Tsinghua University（清华大学）；OpenGVLab（开放通用实验室）；Shanghai AI Laboratory（上海人工智能实验室）；SenseTime Research（商汤科技研究院）；Nanjing University（南京大学）；The Chinese University of Hong Kong（香港中文大学）
Local Margin Restoration for Test-Time Adaptation of Vision-Language Models
Guangzhou University（广州大学）；The Second Affiliated Hospital of Guangzhou University of Chinese Medicine（广州中医药大学第二附属医院）；Jinan University（暨南大学）；Pengcheng Laboratory（鹏城实验室）
DuoTeach: Dual Role Self-Teaching for Coarse-to-Fine Decision Coordination in Vision--Language Models
Department of Computer, North China Electric Power University, Baoding, China（华北电力大学计算机学院）；AIMS Lab, HKUST (GZ), Guangzhou, China（香港科技大学（广州）人工智能实验室）；HKUST, Hong Kong SAR, China（香港科技大学）；School of Information and Intelligent Science, Donghua University, Shanghai, China（东华大学信息与智能科学学院）
Compositional Zero-shot Learning via Progressive Language-based Observations
AI Chip Center for Emerging Smart Systems（新兴智能系统人工智能芯片中心）；The Hong Kong University of Science（香港科学大学）；Zhejiang University College of Computer Science（浙江大学计算机科学学院）
MedObvious: Exposing the Medical Moravec's Paradox in VLMs via Clinical Triage
Mohamed bin Zayed University of Artificial Intelligence, UAE（马尔代夫布扎伊德人工智能大学，阿联酋）；National Institute of Technology, Silchar（西尔CHAR国家理工学院）；Birmingham City University, UK（伯明翰城市大学，英国）
SpatiaLQA: A Benchmark for Evaluating Spatial Logical Reasoning in Vision-Language Models
Zhejiang University（浙江大学）；The University of Sydney（悉尼大学）；ManyCore；State Key Laboratory of Blockchain and Security, Zhejiang University（浙江大学区块链与安全国家重点实验室）；Hangzhou High-Tech Zone (Binjiang) Institute of Blockchain and Data Security（杭州高新技术区（滨江）区块链与数据安全研究院）
How Much Information Can a Vision Token Hold? A Scaling Law for Recognition Limits in VLMs
City University of Hong Kong（香港城市大学）；The Hong Kong Polytechnic University（香港理工大学）；The Hong Kong University of Science and Technology（香港理工大学）；Centre for Artificial Intelligence and Robotics, Chinese Academy of Sciences（中国科学院人工智能与机器人研究院）
