Juan Cao

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

研究方向

AI safety · image and video synthesis · deepfake detection and attribution · fake news and rumor detection · computational social science · multimedia technology

4.5

Generalization through Discrepancy: Leveraging Distributional Fitting Gaps for AI-Generated Image Detection

ICLR 2026Rejected

三作

4.0

Adversaries Fight Back: Revisiting Test-Time Adversarial Defenders of Vision-Language Model under Adaptive Attacks

ICLR 2026Rejected

-1

Accelerating Video Diffusion Transformers with Spatial-Temporal Sparsity

ICLR 2026Desk Rejected

7.0

Understanding and Mitigating Hallucination in Large Vision-Language Models via Modular Attribution and Intervention

ICLR 2025Poster

三作

6.4

From Judgment to Interference: Early Stopping LLM Harmful Outputs via Streaming Content Monitoring

合作者 (14)