9 papers
Geometry-Aware Test-Time Learning for Quantitative Spatial Reasoning
Gege Zhang, Shuaicheng Niu, Gang Dai +2
Quantitative spatial reasoning in visual-language models (VLMs) aims to infer spatial distances and directional relationships among objects in 3D space from a 2D image and a natura…
A Survey on Self-Improving Test-Time Intelligence: Feedback-Driven Adapting, Learning, and Scaling at Inference
Shuaicheng Niu, Guohao Chen, Yaofo Chen +14
The ability of AI systems to improve their behavior during deployment is becoming increasingly important. As inference moves beyond the static execution of a fixed trained model, a…
Guided Trajectory Optimization with Sparse Scaling for Test-Time Diffusion
Gang Dai, Yining Huang, Yiming Xia +2
Test-Time Scaling (TTS) paradigm offers a promising perspective for enhancing the generation performance of diffusion models. However, current solutions largely restrict their sear…
Beyond Isolated Words: Diffusion Brush for Handwritten Text-Line Generation
Gang Dai, Yifan Zhang, Yutao Qin +3
Existing handwritten text generation methods primarily focus on isolated words. However, realistic handwritten text demands attention not only to individual words but also to the r…
Globally Correlation-Aware Hard Negative Generation
Wenjie Peng, Hongxiang Huang, Tianshui Chen +3
Hard negative generation aims to generate informative negative samples that help to determine the decision boundaries and thus facilitate advancing deep metric learning. Current wo…
One-Shot Diffusion Mimicker for Handwritten Text Generation
Gang Dai, Yifan Zhang, Quhui Ke +2
Existing handwritten text generation methods often require more than ten handwriting samples as style references. However, in practical applications, users tend to prefer a handwri…
Diffusion-Driven Data Replay: A Novel Approach to Combat Forgetting in Federated Class Continual Learning
Jinglin Liang, Jin Zhong, Hanlin Gu +6
Federated Class Continual Learning (FCCL) merges the challenges of distributed client learning with the need for seamless adaptation to new classes without forgetting old ones. The…
Disentangling Writer and Character Styles for Handwriting Generation
Gang Dai, Yifan Zhang, Qingfeng Wang +4
Training machines to synthesize diverse handwritings is an intriguing task. Recently, RNN-based methods have been proposed to generate stylized online Chinese characters. However,…
AGTGAN: Unpaired Image Translation for Photographic Ancient Character Generation
Hongxiang Huang, Daihui Yang, Gang Dai +7
The study of ancient writings has great value for archaeology and philology. Essential forms of material are photographic characters, but manual photographic character recognition…