Duc Vu

πŸ‘‹ Hi! I'm Duc. Nice to meet y'all! I am an incoming Fall 2026 PhD student at Rutgers University under the supervision of Distinguished Prof. Dimitris N. Metaxas. I am currently a predoctoral researcher at Qualcomm AI Research πŸ“‘, where I am fortunate to be supervised by Dr. Anh Tran. Prior to this, I received my B.S. in Data Science & Statistics and B.A. in Economics from πŸŽ“ Miami University - Oxford in the U.S. (2024).

πŸ”  Open to Summer 2027 Research Internships or research collaborations. Let’s create impactful work together!  πŸš€

Research

My research interests include physics-aware and object-interaction video generation, with the goal of developing generative models that can better understand, capture, and simulate real-world physical dynamics. Previously, my work focused on diffusion-based generative models, with an emphasis on diffusion distillation and efficient one-step generation techniques for practical applications such as video enhancement and image inpainting. I am particularly interested in methods that maintain high generation quality while reducing computational cost, making generative AI more scalable, efficient, and accessible. I have also worked on adversarial defenses to support the safe and responsible deployment of generative video models.

Physics-Aware Video Gen One/Few-Step Diffusion Diffusion Distillation Video Enhancement Image Inpainting Adversarial Defense

Selected Publications

* denotes equal contribution

Cross-Space Distillation Preview
Cross-Space Distillation: Teaching One-Step Students with Modern Diffusion Teachers.
ECCV 2026

We formalize Cross-Space Distillation and introduce Bridge, a lightweight latent-space interface that makes standard one-step distillation possible across mismatched resolutions, VAEs, architectures, and diffusion/flow paradigms.

Anti-I2V: Safeguarding your photos from malicious image-to-video generation.

A memory-efficient adversarial attack against diverse image-to-video diffusion models, enabled by robust dual-space perturbation optimization.

InverFill: One-Step Inversion for Enhanced Few-Step Diffusion Inpainting.

A highly efficient one-step inversion diffusion network for high-quality few-step image inpainting.

Improved Training Technique for Shortcut Models.
NeurIPS 2025

iSM resolves five major shortcut-model flaws with dynamic guidance, wavelet loss, sOT, and Twin EMA, yielding markedly better image generation.

SwiftBrush v2: Make Your One-step Diffusion Model Better Than Its Teacher.

An improved SwiftBrush version that makes the one-step diffusion student beats its multi-step teacher.

EFHQ: Multi-purpose ExtremePose-Face-HQ dataset.

A high-quality dataset centered on extreme pose faces, supporting face synthesis, reenactment, recognition benchmarking, and more.

Selected Preprints

* denotes equal contribution

VideoDrift: Plug-and-Play Video Refinement for Diffusion Models via KV-Anchored Attention.
Ngan Nguyen, Duc Vu, Trong-Tung Nguyen, Phuc Hong Lai, Cuong Pham, Anh Tran
Under Review 2026

A plug-and-play, backbone-agnostic video enhancer with low compute: each frame requires only inversion plus one generator pass, making it easy to deploy on outputs from diverse T2V systems.