9 papers
AVA-Encoder: Towards Agent-Native Video Representation Learning
Chuyue Li, Jinpeng Yu, Haozhe Wang +7
Agentic video creation systems remain far from cinematic-grade quality: their underlying models struggle to plan how story, characters, visuals, and sound fit together, and high-qu…
What to Edit Next: Visually Aligned Image-Editing Follow-Up Suggestions in Conversational Systems
Zhijing Zhang, Jinpeng Yu, Xin Song +6
Conversational assistants increasingly recommend follow-up edits to help users continue a task. Existing systems primarily target text-only interactions, leaving image-creation con…
Rethinking Classifier-Free Guidance in On-Policy Diffusion Distillation
Bingnan Li, Haozhe Wang, Haozhong Xiong +5
On-policy distillation (OPD) adapts diffusion models by querying a teacher along trajectories generated by the current student, but how it should behave under classifier-free guida…
DLLG: Dynamic Logit-Level Gating of LLM Experts
Bingnan Li, Zhaoyang Zhang, Xiaoze Liu +6
Leveraging multiple specialized LLMs can combine complementary strengths, but existing approaches trade adaptability for stability: routing commits prematurely, heuristic ensemblin…
CyCLeGen: Cycle-Consistent Layout Prediction and Image Generation in Vision Foundation Models
Xiaojun Shan, Haoyu Shen, Yucheng Mao +5
We present CyCLeGen, a unified vision-language foundation model capable of both image understanding and image generation within a single autoregressive framework. Unlike existing v…
OverLayBench: A Benchmark for Layout-to-Image Generation with Dense Overlaps
Bingnan Li, Chen-Yu Wang, Haiyang Xu +7
Despite steady progress in layout-to-image generation, current methods still struggle with layouts containing significant overlap between bounding boxes. We identify two primary ch…
YOLO-Count: Differentiable Object Counting for Text-to-Image Generation
Guanning Zeng, Xiang Zhang, Zirui Wang +4
We propose YOLO-Count, a differentiable open-vocabulary object counting model that tackles both general counting challenges and enables precise quantity control for text-to-image (…
Generalize or Detect? Towards Robust Semantic Segmentation Under Multiple Distribution Shifts
Zhitong Gao, Bingnan Li, Mathieu Salzmann +1
In open-world scenarios, where both novel classes and domains may exist, an ideal segmentation model should detect anomaly classes for safety and generalize to new domains. However…
Gradient-Map-Guided Adaptive Domain Generalization for Cross Modality MRI Segmentation
Bingnan Li, Zhitong Gao, Xuming He
Cross-modal MRI segmentation is of great value for computer-aided medical diagnosis, enabling flexible data acquisition and model generalization. However, most existing methods hav…