publications

(*) denotes equal contribution

2024

  1. ICLROral
    Improved Techniques for Training Consistency Models
    Yang Song, and Prafulla Dhariwal
    In the 12th International Conference on Learning Representations, 2024.
    Oral Presentation [Top 1.2%]
  2. ICLR
    Diffusion Posterior Sampling for Linear Inverse Problem Solving: A Filtering Perspective
    Zehao Dou, and Yang Song
    In the 12th International Conference on Learning Representations, 2024.

2023

  1. ICML
    Consistency Models
    In the 40th International Conference on Machine Learning, 2023.

2022

  1. ACM
    Diffusion Models: A Comprehensive Survey of Methods and Applications
    Ling Yang, Zhilong Zhang, Yang Song, Shenda Hong, Runsheng Xu, Yue Zhao, Wentao Zhang, Bin Cui, and Ming-Hsuan Yang
    ACM Computing Surveys
  2. Thesis
    Learning to Generate Data by Estimating Gradients of the Data Distribution
    Yang Song
    Stanford University
  3. ICLR
    Solving Inverse Problems in Medical Imaging with Score-Based Generative Models
    Yang Song*, Liyue Shen*, Lei Xing, and Stefano Ermon
    In the 10th International Conference on Learning Representations, 2022. Abridged in the NeurIPS 2021 Workshop on Deep Learning and Inverse Problems.
  4. ICLR
    SDEdit: Guided Image Synthesis and Editing with Stochastic Differential Equations
    Chenlin Meng, Yutong He, Yang Song, Jiaming Song, Jiajun WuJun-Yan Zhu, and Stefano Ermon
    In the 10th International Conference on Learning Representations, 2022.
  5. ICLROral
    GeoDiff: A Geometric Diffusion Model for Molecular Conformation Generation
    Minkai Xu, Lantao Yu, Yang Song, Chence Shi, Stefano Ermon, and Jian Tang
    In the 10th International Conference on Learning Representations, 2022.
    Oral Presentation [Top 1.6%]
  6. AISTATSOral
    Density Ratio Estimation via Infinitesimal Classification
    Kristy Choi, Chenlin Meng, Yang Song, and Stefano Ermon
    In the 25th International Conference on Artificial Intelligence and Statistics, 2022.
    Oral Presentation [Top 2.6%]

2021

  1. arXiv
    Score-Based Generative Classifiers
    Roland S. Zimmermann, Lukas Schott, Yang Song, Benjamin A. Dunn, and David A. Klindt
    In arXiv preprint arXiv:2110.00473. Abridged in the NeurIPS 2021 Workshop on Deep Generative Models and Downstream Applications.
  2. NeurIPSSpotlight
    Maximum Likelihood Training of Score-Based Diffusion Models
    Yang Song*, Conor Durkan*, Iain Murray, and Stefano Ermon
    In the 35th Conference on Neural Information Processing Systems, 2021.
    Spotlight Presentation [top 3%]
  3. NeurIPS
    Estimating High Order Gradients of the Data Distribution by Denoising
    Chenlin Meng, Yang Song, Wenzhe Li, and Stefano Ermon
    In the 35th Conference on Neural Information Processing Systems, 2021.
  4. NeurIPS
    Pseudo-Spherical Contrastive Divergence
    Lantao Yu, Jiaming Song, Yang Song, and Stefano Ermon
    In the 35th Conference on Neural Information Processing Systems, 2021.
  5. NeurIPS
    Imitation with Neural Density Models
    Kuno Kim, Akshat Jindal, Yang Song, Jiaming Song, Yanan Sui, and Stefano Ermon
    In the 35th Conference on Neural Information Processing Systems, 2021.
  6. NeurIPS
    CSDI: Conditional Score-based Diffusion Models for Probabilistic Time Series Imputation
    Yusuke Tashiro, Jiaming Song, Yang Song, and Stefano Ermon
    In the 35th Conference on Neural Information Processing Systems, 2021.
  7. ICML
    Accelerating Feedforward Computation via Parallel Nonlinear Equation Solving
    Yang Song, Chenlin Meng, Renjie Liao, and Stefano Ermon
    In the 38th International Conference on Machine Learning, 2021.
  8. Book Chapter
    How to Train Your Energy-Based Models
    Yang Song, and Diederik P. Kingma
    In “Probabilistic Machine Learning: Advanced Topics” by Kevin P. Murphy.
  9. ICLROralAward
    Score-Based Generative Modeling through Stochastic Differential Equations
    In the 9th International Conference on Learning Representations, 2021.
    Outstanding Paper Award
  10. ICLROral
    Improved Autoregressive Modeling with Distribution Smoothing
    Chenlin Meng, Jiaming Song, Yang Song, Shengjia Zhao, and Stefano Ermon
    In the 9th International Conference on Learning Representations, 2021.
    Oral Presentation [top 1.8%]
  11. ICLR
    Learning Energy-Based Models by Diffusion Recovery Likelihood
    Ruiqi Gao, Yang SongBen PooleYing Nian Wu, and Diederik P. Kingma
    In the 9th International Conference on Learning Representations, 2021.
  12. ICLR
    Anytime Sampling for Autoregressive Models via Ordered Autoencoding
    Yilun Xu, Yang Song, Sahaj Garg, Linyuan Gong, Rui Shu, Aditya Grover, and Stefano Ermon
    In the 9th International Conference on Learning Representations, 2021.

2020

  1. NeurIPS
    Improved Techniques for Training Score-Based Generative Models
    Yang Song, and Stefano Ermon
    In the 34th Conference on Neural Information Processing Systems, 2020.
  2. NeurIPS
    Diversity can be Transferred: Output Diversification for White- and Black-box Attacks
    Yusuke Tashiro, Yang Song, and Stefano Ermon
    In the 34th Conference on Neural Information Processing Systems, 2020.
  3. NeurIPS
    Autoregressive Score Matching
    Chenlin Meng, Lantao Yu, Yang Song, Jiaming Song, and Stefano Ermon
    In the 34th Conference on Neural Information Processing Systems, 2020.
  4. NeurIPS
    Efficient Learning of Generative Models via Finite-Difference Score Matching
    Tianyu Pang, Kun Xu, Chongxuan Li, Yang SongStefano Ermon, and Jun Zhu
    In the 34th Conference on Neural Information Processing Systems, 2020.
  5. ICML
    Training Deep Energy-Based Models with f-Divergence Minimization
    Lantao Yu, Yang Song, Jiaming Song, and Stefano Ermon
    In the 37th International Conference on Machine Learning, 2020.
  6. AISTATS
    Gaussianization Flows
    Chenlin Meng*, Yang Song*, Jiaming Song, and Stefano Ermon
    In the 23rd International Conference on Artificial Intelligence and Statistics, 2020.
  7. AISTATS
    Permutation Invariant Graph Generation via Score-Based Generative Modeling
    Chenhao Niu, Yang Song, Jiaming Song, Shengjia Zhao, Aditya Grover, and Stefano Ermon
    In the 23rd International Conference on Artificial Intelligence and Statistics, 2020.

2019

  1. arXiv
    Unsupervised Out-of-Distribution Detection with Batch Normalization
    Jiaming Song, Yang Song, and Stefano Ermon
    In Technical report (10/21/2019).
  2. OpenReview
    Towards Certified Defense for Unrestricted Adversarial Attacks
    Shengjia Zhao, Yang Song, and Stefano Ermon
    In Technical report, 2019.
  3. NeurIPSOral
    Generative Modeling by Estimating Gradients of the Data Distribution
    Yang Song, and Stefano Ermon
    In the 33rd Conference on Neural Information Processing Systems, 2019.
    Oral Presentation [top 0.5%]
  4. NeurIPS
    MintNet: Building Invertible Neural Networks with Masked Convolutions
    Yang Song*, Chenlin Meng*, and Stefano Ermon
    In the 33rd Conference on Neural Information Processing Systems, 2019.
  5. NeurIPS
    Efficient Graph Generation with Graph Recurrent Attention Networks
    Renjie Liao, Yujia Li, Yang Song, Shenlong Wang, William L. HamiltonDavid DuvenaudRaquel Urtasun, and Richard Zemel
    In the 33rd Conference on Neural Information Processing Systems, 2019.
  6. UAIOral
    Sliced Score Matching: A Scalable Approach to Density and Score Estimation
    Yang Song*, Sahaj Garg*, Jiaxin Shi, and Stefano Ermon
    In the 35th Conference on Uncertainty in Artificial Intelligence, 2019.
    Oral Presentation [top 8.7%]

2018

  1. NeurIPS
    Constructing Unrestricted Adversarial Examples with Generative Models
    Yang Song, Rui Shu, Nate Kushman, and Stefano Ermon
    In the 32nd Conference on Neural Information Processing Systems, 2018.
  2. ICML
    Accelerating Natural Gradient with Higher-Order Invariance
    Yang Song, Jiaming Song, and Stefano Ermon
    In the 35th International Conference on Machine Learning, 2018.
  3. ICLR
    PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples
    Yang Song, Taesup Kim, Sebastian NowozinStefano Ermon, and Nate Kushman
    In the 6th International Conference on Learning Representations, 2018.

2016

  1. NeurIPS
    Kernel Bayesian Inference with Posterior Regularization
    Yang SongJun Zhu, and Yong Ren
    In the 30th Conference on Neural Information Processing Systems, 2016.
  2. NeurIPS
    Stochastic Gradient Geodesic MCMC Methods
    Chang Liu, Jun Zhu, and Yang Song
    In the 30th Conference on Neural Information Processing Systems, 2016.
  3. ICML
    Training Deep Neural Networks via Direct Loss Minimization
    In the 33rd International Conference on Machine Learning, 2016.
  4. AAAIOral
    Bayesian Matrix Completion via Adaptive Relaxed Spectral Regularization
    Yang Song, and Jun Zhu
    In the 30th AAAI Conference on Artificial Intelligence, 2016.