Ning Yu 于宁

About me

I am a research scientist at Netflix Eyeline Studios, leading efforts in visual and multimodal generative AI. My research aspirations lie in controllable and customizable recreation of the world via generative AI, based on representations of multimodal contents and interactions between users and models.

I hold a joint Ph.D. from the University of Maryland and Max Planck Institute for Informatics where I was supervised by Larry Davis and Mario Fritz. I previously worked with Salesforce, NVIDIA, and Adobe.

My selected works include UniControl, X-InstructBLIP, ULIP-2, LayoutDETR, GlueGen, HIVE, Contrast and Attention GANs, Scalable GAN Fingerprints, Artificial GAN Fingerprints, GAN Fingerprints, Inclusive GAN, and Texture Mixer. My research was featured in media press such as Salesforce Newsletter, AlmostHuman, AIera, JiangMen, QbitAI, innovations report, and FoodPlus.

I am constantly dedicated to academic services, such as serving as (Lead) Area Chair for CVPR, ICCV, ECCV, NeurIPS, ICLR, AAAI, AISTATS, WACV, and BMVC.

I am a recipient of the CSAW Europe Best Paper Finalist, Twitch (Amazon) Research Fellowship, Microsoft Young Fellowship, Qualcomm Innovation Fellowship Finalist x2, and SPIE Best Student Paper Finalist.

News

[2024/09] Text2Data is accepted to AAAI 2025. The code is released.
[2024/09] T2Vs Meet VLMs is accepted to NeurIPS 2024. The code and data card are released.
[2024/09] Shadowcast is accepted to NeurIPS 2024. The code is released.
[2024/09] FastTrack is accepted to EMNLP 2024.
[2024/08] xGen-MM (BLIP-3) is accepted to ECCV Workshop on EVAL-FoMo 2024. The code, model cards, and data cards are released.
[2024/08] Inside the Black Box is accepted to ECAI 2024.
[2024/07] DifFRelight is accepted to SIGGRAPH Asia 2024.
[2024/07] Vision-Language Models for Food Nutrient Screening is accepted to Advanced Science 2024.
[2024/07] LayoutDETR and X-InstructBLIP are accepted to ECCV 2024.
[2024/05] RigorLLM and C-RAG are accepted to ICML 2024.
[2024/04] LLMs in Food Science is accepted to Trends in Food Science & Technology 2024.
[2024/03] SimSCOOD is accepted to NAACL 2024. The code is released.
[2024/02] HIVE and ULIP-2 are accepted to CVPR 2024.
[2024/01] Model-Agnostic Hierarchical Attention for 3D Object Detection is accepted to ICRA 2024.
[2023/11] AnchMark is accepted to NeurIPS Workshop on Regulatable ML 2023.
[2023/09] UniControl is accepted to NeurIPS 2023. The code is released.
[2023/07] GlueGen is accepted to ICCV 2023. The code is released.
[2023/05] DE-FAKE is accepted to CCS 2023 as in the CSAW Europe best paper finalist. The code is released.
[2023/04] Generated Graph Detection is accepted to ICML 2023. The code is released.
[2023/04] SPAD and Prototype LTR are accepted to IJCAI 2023.
[2023/04] RoSteALS is accepted to CVPR Workshop on Media Forensics 2023. The code is released.
[2023/02] Mask-free OVIS and Cont-Steal are accepted to CVPR 2023.
[2022/10] UnGANable is accepted to USENIX 2023. The code is released.
[2022/08] FcF Image Inpainting is accepted to WACV 2023. The code is released.
[2022/07] I am one of the top reviewers and serve as a session chair for ICML 2022.
[2022/07] RepMix is accepted to ECCV 2022. The code is released.
[2022/05] Deep Learning for Food Nutrient Estimation is accepted to Food Chemistry 2022.
[2022/04] Multi-Exit Privacy is accepted to CCS 2022. The code is released.
[2022/04] The code for RelaxLoss is released.
[2022/01] The code for Scalable GAN Fingerprints is released.
[2022/01] Scalable GAN Fingerprints and RelaxLoss are accepted to ICLR 2022 as Spotlight.
[2021/10] VIDNet is accepted to BMVC 2021. The code is released.
[2021/08] I defended my Ph.D. The dissertation about Human-Centric Deep Generative Models is online.
[2021/08] Computer Vision for Food Nutrient Estimation is accepted to Food Chemistry 2021.
[2021/07] Artificial GAN Fingerprints is accepted to ICCV 2021 as Oral. The code is released.
[2021/07] Dual Contrastive Loss and Attention for GANs is accepted to ICCV 2021.
[2021/05] Computer Vision for Food Nutrient Prediction is accepted to Food Research International 2021.
[2021/04] Re-Synthesis for Deepfake Detection is accepted to IJCAI 2021. The code is released.
[2021/03] AI for Food Nutrient Estimation is accepted to Journal of Food Composition and Analysis 2021.
[2021/02] Hijack-GAN is accepted to CVPR 2021. The code is released.
[2020/08] Class Balanced Experts for Long-Tailed Recognition is accepted to GCPR 2020. The code is released.
[2020/06] Inclusive GAN is accepted to ECCV 2020. The code is released.
[2020/03] GAN-Leaks is accepted to CCS 2020. The code is released.

Publications