4 papers
TC-AE: Unlocking Token Capacity for Deep Compression Autoencoders
Teng Li, Ziyuan Huang, Cong Chen +5
We propose TC-AE, a ViT-based architecture for deep compression autoencoders. Existing methods commonly increase the channel number of latent representations to maintain reconstruc…
UniFork: Exploring Modality Alignment for Unified Multimodal Understanding and Generation
Teng Li, Quanfeng Lu, Lirui Zhao +5
Unified image understanding and generation has emerged as a promising paradigm in multimodal artificial intelligence. Despite recent progress, the optimal architectural design for…
Task-Oriented Communication for Vehicle-to-Infrastructure Cooperative Perception
Jiawei Shao, Teng Li, Jun Zhang
Vehicle-to-infrastructure (V2I) cooperative perception plays a crucial role in autonomous driving scenarios. Despite its potential to improve perception accuracy and robustness, th…
Spatial-Aware Latent Initialization for Controllable Image Generation
Wenqiang Sun, Teng Li, Zehong Lin +1
Recently, text-to-image diffusion models have demonstrated impressive ability to generate high-quality images conditioned on the textual input. However, these models struggle to ac…