8 papers
VideoJev: System One Model for Video Understanding and Interaction
Yuhan Zhu, Xiangyu Zeng, Qingyi Si +15
Visual agents must turn video observations into timely decisions to track progress and choose what to do next. Building on the emerging System One paradigm, we introduce VideoJev,…
VideoChat3: Fully Open Video MLLM for Efficient and Generalist Video Understanding
Xinhao Li, Yuhan Zhu, Xiangyu Zeng +24
Recent advances in video understanding have spanned motion, long video, and streaming interaction, driving this field toward real-world applications. Despite this progress, current…
TimeLens2: Generalist Video Temporal Grounding with Multimodal LLMs
Yuhan Zhu, Changlian Ma, Xiangyu Zeng +12
Video multimodal large language models (MLLMs) can describe what happens in a video, but rarely identify when the supporting evidence occurs. We study generalist video temporal gro…
FlowErase-RL: Rethinking Concept Erasure as Reward Optimization in Flow Matching Models
Yi Sun, Zhiqi Zhang, Xinhao Zhong +5
Recent advances in flow matching models have significantly improved text-to-image generation quality, but also introduce growing safety risks due to the generation of harmful or un…
SimInsert: Seamless Video Object Insertion via Regional Sparse Attention Fusion
Xinyu Chen, Yuyi Qian, Jiang Lin +9
Video object insertion requires ensuring spatio-temporal coherence and interactive realism, extending far beyond simple content placement. However, current approaches are often hin…
Video-o3: Native Interleaved Clue Seeking for Long Video Multi-Hop Reasoning
Xiangyu Zeng, Zhiqiu Zhang, Yuhan Zhu +12
Existing multimodal large language models for long-video understanding predominantly rely on uniform sampling and single-turn inference, limiting their ability to identify sparse y…
FreeControl: Efficient, Training-Free Structural Control via One-Step Attention Extraction
Jiang Lin, Xinyu Chen, Song Wu +7
Controlling the spatial and semantic structure of diffusion-generated images remains a challenge. Existing methods like ControlNet rely on handcrafted condition maps and retraining…
Region-to-Region: Enhancing Generative Image Harmonization with Adaptive Regional Injection
Zhiqiu Zhang, Dongqi Fan, Mingjie Wang +3
The goal of image harmonization is to adjust the foreground in a composite image to achieve visual consistency with the background. Recently, latent diffusion model (LDM) are appli…