← Search

Lei Gan

2 accepted papers

2024

BayesDiff: Estimating Pixel-wise Uncertainty in Diffusion via Bayesian Inference

ICLR 2024poster

Diffusion models have impressive image generation capability, but low-quality generations still exist, and their identification remains challenging due to the lack of a proper sample-wise metric. To address this, we propose BayesDiff, a pixel-wise uncertainty estimator for generations from diffusion…

2024

Dissecting Dissonance: Benchmarking Large Multimodal Models Against Self-Contradictory Instructions

ECCV 2024poster

"Large multimodal models (LMMs) excel in adhering to human instructions. However, self-contradictory instructions may arise due to the increasing trend of multimodal interaction and context length, which is challenging for language beginners and vulnerable populations. We introduce the Self-Contradi…