Schedule
October 20, 2025 | 8:00 AM – 5:15 PM | Room 304-A
8:00 – 8:30 AM
Welcome
8:30 – 9:15 AM
Invited Speaker: Zhuang Liu
9:15 – 10:00 AM
Oral Session 1
How to Train your Text-to-Image Model: Evaluating Design Choices for Synthetic Training Captions
Generating Fine Details of Entity Interactions
Role Bias in Text-to-Image Diffusion Models: Diagnosing and Mitigating Compositional Failures through Intermediate Decomposition
10:00 – 10:45 AM
Invited Speaker: Sara Beery
11:00 AM – 12:00 PM
Poster Session
ExHall II
12:00 – 1:30 PM
Lunch Break
1:30 – 2:15 PM
Oral Session 2
Synthetic Captions for Open-Vocabulary Zero-Shot Segmentation
A CLIP-Powered Framework for Robust and Generalizable Data Selection
VILA²: Towards VLM Augmentation via Self-Improvement
2:15 – 3:00 PM
Invited Speaker: Andrew Owens
3:00 – 3:45 PM
Oral Session 3
SYM3D: Canonicalizing Triplanes via Symmetry for Single-View 3D Learning
Objaverse++: Curated 3D Object Dataset with Quality Annotations
ControlTac: Force- and Position-Controlled Tactile Data Augmentation with a Single Reference Image
3:45 – 4:30 PM
Invited Speaker: Phillip Isola
4:30 – 5:15 PM