5 papers
Think in Strokes, Not Pixels: Process-Driven Image Generation via Interleaved Reasoning
Lei Zhang, Junjiao Tian, Zhipeng Fan +9
Humans paint images incrementally: they plan a global layout, sketch a coarse draft, inspect, and refine details, and most importantly, each step is grounded in the evolving visual…
Directional Gradient Projection for Robust Fine-Tuning of Foundation Models
Chengyue Huang, Junjiao Tian, Brisa Maneechotesuwan +2
Robust fine-tuning aims to adapt large foundation models to downstream tasks while preserving their robustness to distribution shifts. Existing methods primarily focus on constrain…
Token-Shuffle: Towards High-Resolution Image Generation with Autoregressive Models
Xu Ma, Peize Sun, Haoyu Ma +22
Autoregressive (AR) models, long dominant in language generation, are increasingly applied to image synthesis but are often considered less competitive than Diffusion-based models.…
Grounding Descriptions in Images informs Zero-Shot Visual Recognition
Shaunak Halbe, Junjiao Tian, K J Joseph +4
Vision-language models (VLMs) like CLIP have been cherished for their ability to perform zero-shot visual recognition on open-vocabulary concepts. This is achieved by selecting the…
Rethinking Weight Decay for Robust Fine-Tuning of Foundation Models
Junjiao Tian, Chengyue Huang, Zsolt Kira
Modern optimizers such as AdamW, equipped with momentum and adaptive learning rate, are designed to escape local minima and explore the vast parameter space. This exploration is be…