Qianlong Xiang — accepted AI-conference papers

2026

TINA: Text-Free Inversion Attack for Unlearned Text-to-Image Diffusion Models

Qianlong Xiang, Miao Zhang, Haoyu Zhang, Kun Wang, Junhui Hou, Liqiang Nie

CVPR 2026

Although text-to-image diffusion models exhibit remarkable generative power, concept erasure techniques are essential for their safe deployment to prevent the creation of harmful content.This has fostered a dynamic interplay between the development of erasure defenses and the adversarial probes desi

Cited by 0 Source ↗code ↗Scholar ↗

2025

DKDM: Data-Free Knowledge Distillation for Diffusion Models with Any Architecture

Qianlong Xiang, Miao Zhang, Yuzhang Shang, Jianlong Wu, Yan Yan, Liqiang Nie

CVPR 2025poster

Diffusion models (DMs) have demonstrated exceptional generative capabilities across various domains, including image, video, and so on. A key factor contributing to their effectiveness is the high quantity and quality of data used during training. However, mainstream DMs now consume increasingly lar…

Cited by 8 Source ↗PDF ↗code ↗Scholar ↗