Shiguang Shan

Full Professor@Institute of Computing Technology, Chinese Academy of Sciences·中国·OpenReview

研究方向

AI safety · large models safety evaluation · Transformers · Foundation models · Self-supervised learning · weakly-supervised learning · Face anti-spoofing · present attack detection · Lip reading · visual speech recognition · Pedestrian re-identification · Deep learning · CNNs · domain adaptation · Facial expression recognition · facial action unit detection · affective computing · Face recognition · face detection · face verification

Shiguang Shan

Plan-R1: Safe and Feasible Trajectory Planning as Language Modeling

Uniform Discrete Diffusion with Metric Path for Video Generation

HumanPCR: Probing MLLM Capabilities in Diverse Human-Centric Scenes

Multi-Level CLIP Transfer for Open-Vocabulary Object Detection

Seeing Like Humans: Task-Driven Token Reduction for Accelerated ViT in Robotic Navigation

Jodi: Unification of Visual Generation and Understanding via Joint Modeling

Distribution-Aware Synergistic Evolution for Few-shot Discrimination and Generation

A Dual-Branch Disentanglement Diffusion for ID-Attribute Conditional Face Generation

Towards Non-destructive Privacy Protection for LVLMs via node-level localized editing

ProtInvTree: Deliberate Protein Inverse Folding with Reward-guided Tree Search

Autoregressive Video Generation without Vector Quantization

Revisiting Logit Distributions for Reliable Out-of-Distribution Detection

CtrLoRA: An Extensible and Efficient Framework for Controllable Image Generation

Dysca: A Dynamic and Scalable Benchmark for Evaluating Perception Ability of LVLMs

un$^2$CLIP: Improving CLIP's Visual Detail Capturing Ability via Inverting unCLIP

Evaluating the Quality of Hallucination Benchmarks for Large Vision-Language Models

MATS: An Audio Language Model under Text-only Supervision

Enhancing Zero-shot OOD Detection with Pre-trained Multimodal Foundation Models

Not Only Vision: Evolve Visual Speech Recognition via Peripheral Information

CM^2: Cross-Modal Contextual Modeling for Audio-Visual Speech Enhancement

Semantic or Covariate? A Study on the Intractable Case of Out-of-Distribution Detection