4 papers
BEVDiffuser: Plug-and-Play Diffusion Model for BEV Denoising with Ground-Truth Guidance
Xin Ye, Burhaneddin Yaman, Sheng Cheng +3
Bird's-eye-view (BEV) representations play a crucial role in autonomous driving tasks. Despite recent advancements in BEV generation, inherent noise, stemming from sensor limitatio…
Latent Space Energy-based Neural ODEs
Sheng Cheng, Deqian Kong, Jianwen Xie +3
This paper introduces novel deep dynamical models designed to represent continuous-time sequences. Our approach employs a neural emission model to generate each data point in the t…
Precision or Recall? An Analysis of Image Captions for Training Text-to-Image Generation Model
Sheng Cheng, Maitreya Patel, Yezhou Yang
Despite advancements in text-to-image models, generating images that precisely align with textual descriptions remains challenging due to misalignment in training data. In this pap…
TripletCLIP: Improving Compositional Reasoning of CLIP via Synthetic Vision-Language Negatives
Maitreya Patel, Abhiram Kusumba, Sheng Cheng +4
Contrastive Language-Image Pretraining (CLIP) models maximize the mutual information between text and visual modalities to learn representations. This makes the nature of the train…