14 papers
Uncertainty-Aware Consistency Distillation for Few-Step Video Generation
Lingyu Liu, Yaxiong Wang, Li Zhu +1
We study few-step video generation, i.e., distilling a multi-step video generator, which typically requires tens of sampling steps, incurring substantial latency and compute, into…
Before Thinking, Learn to Decide: Proactive Routing for Efficient Visual Reasoning
Yinan Zhou, Haokun Lin, Yichen Wu +7
Large multimodal models have achieved strong reasoning on complex visual tasks, but their inference efficiency is often restricted by long chains of thought. A promising solution i…
StepGuard: Guarding Web Navigation via Single-Step Calibration
Zhihao Cui, Yuchen Zhang, Xiyang Sun +6
Web navigation requires agents to follow natural language goals, interact with web pages, and produce accurate answers. While recent advances leverage vision-language models and re…
Can Video Diffusion Models Predict Past Frames? Bidirectional Cycle Consistency for Reversible Interpolation
Lingyu Liu, Yaxiong Wang, Li Zhu +1
Video frame interpolation aims to synthesize realistic intermediate frames between given endpoints while adhering to specific motion semantics. While recent generative models have…
Look, Compare and Draw: Differential Query Transformer for Automatic Oil Painting
Lingyu Liu, Yaxiong Wang, Li Zhu +2
This work introduces a new approach to automatic oil painting that emphasizes the creation of dynamic and expressive brushstrokes. A pivotal challenge lies in mitigating the duplic…
Cultivating Forensic Reasoning for Generalizable Multimodal Manipulation Detection
Yuchen Zhang, Yaxiong Wang, Kecheng Han +4
Recent advances in generative AI have significantly enhanced the realism of multimodal media manipulation, thereby posing substantial challenges to manipulation detection. Existing…
Minimizing the Pretraining Gap: Domain-aligned Text-Based Person Retrieval
Shuyu Yang, Yaxiong Wang, Yongrui Li +2
In this work, we focus on text-based person retrieval, which identifies individuals based on textual descriptions. Despite advancements enabled by synthetic data for pretraining, a…
Towards Scalable Video Anomaly Retrieval: A Synthetic Video-Text Benchmark
Shuyu Yang, Yilun Wang, Yaxiong Wang +2
Video anomaly retrieval aims to localize anomalous events in videos using natural language queries to facilitate public safety. However, existing datasets suffer from severe limita…
The Coherence Trap: When MLLM-Crafted Narratives Exploit Manipulated Visual Contexts
Yuchen Zhang, Yaxiong Wang, Yujiao Wu +3
The detection and grounding of multimedia manipulation has emerged as a critical challenge in combating AI-generated disinformation. While existing methods have made progress in re…
Scale Up Composed Image Retrieval Learning via Modification Text Generation
Yinan Zhou, Yaxiong Wang, Haokun Lin +3
Composed Image Retrieval (CIR) aims to search an image of interest using a combination of a reference image and modification text as the query. Despite recent advancements, this ta…
Every Painting Awakened: A Training-free Framework for Painting-to-Animation Generation
Lingyu Liu, Yaxiong Wang, Li Zhu +1
We introduce a training-free framework specifically designed to bring real-world static paintings to life through image-to-video (I2V) synthesis, addressing the persistent challeng…
Generating Attribution Reports for Manipulated Facial Images: A Dataset and Baseline
Jingchun Lian, Lingyu Liu, Yaxiong Wang +4
Existing facial forgery detection methods typically focus on binary classification or pixel-level localization, providing little semantic insight into the nature of the manipulatio…
Beyond Walking: A Large-Scale Image-Text Benchmark for Text-based Person Anomaly Search
Shuyu Yang, Yaxiong Wang, Li Zhu +1
Text-based person search aims to retrieve specific individuals across camera networks using natural language descriptions. However, current benchmarks often exhibit biases towards…
DOGR: Towards Versatile Visual Document Grounding and Referring
Yinan Zhou, Yuxin Chen, Haokun Lin +6
With recent advances in Multimodal Large Language Models (MLLMs), grounding and referring capabilities have gained increasing attention for achieving detailed understanding and fle…