Cheng Shi

10

Papers

17

Total Citations

Papers (10)

Rethinking Query-based Transformer for Continual Image Segmentation

A Graph Dynamics Prior for Relational Inference

Vision Function Layer in Multimodal LLMs

Sim-DETR: Unlock DETR for Temporal Sentence Grounding

LoGoPrompt: Synthetic Text Images Can Be Good Visual Prompts for Vision-Language Models

Contrastive Grouping With Transformer for Referring Image Segmentation

EdaDet: Open-Vocabulary Object Detection Using Early Dense Alignment

Spatial and Visual Perspective-Taking via View Rotation and Relation Reasoning for Embodied Reference Understanding

Robust Reinforcement Learning via Adversarial training with Langevin Dynamics

Free-Bloom: Zero-Shot Text-to-Video Generator with LLM Director and LDM Animator