Update: I have graduated and this website is inactive. My current website can be found here.

Congzheng Song / 宋丛峥

Curriculum Vitae [pdf]


Email: cs2296[at]cornell.edu
Links: [Google scholar ] [Github ] [Linkedin ]

About Me

Hello! I am a Computer Science Ph.D. candidate at Cornell University (physically located at Cornell Tech) working with Prof. Vitaly Shmatikov. My current research interests are security & privacy issues in machine learning. I completed my bachelor's degree at Emory University, where I worked closely with Prof. Ymir Vigfusson and Prof. Lee Cooper on some fun real world deep learning application projects.

Industrial Experience

Applied scientist intern at Amazon, Summer 2020
Research intern at Google Brain, Fall 2019
Research intern at Petuum Inc, Summer 2019


(* indicates equal contribution)

  1. You Autocomplete Me: Poisoning Vulnerabilities in Neural Code Completion [pdf]
    R.Schuster, C.Song, E. Tromer, V.Shmatikov
    To appear in USENIX Security Symposium, 2021

  2. Adversarial Semantic Collisions [pdf][code]
    C.Song, A.Rush, V.Shmatikov
    In Conference on Empirical Methods in Natural Language Processing (EMNLP), 2020

  3. Information Leakage in Embedding Models [pdf][code]
    C.Song, A.Raghunathan
    In ACM Conference on Computer and Communications Security (CCS), 2020

  4. Generalized Zero-Shot Text Classification for ICD Coding [pdf][code]
    C.Song, S.Zhang, N.Sadoughi, P.Xie, E.P.Xing
    In International Joint Conference on Artificial Intelligence (IJCAI), 2020

  5. Membership Encoding for Deep Learning [pdf]
    C.Song, R.Shokri
    In ACM ASIA Conference on Computer and Communications Security (AsiaCCS), 2020

  6. Overlearning Reveals Sensitive Attributes [pdf][code][slides]
    C.Song, V.Shmatikov
    In International Conference on Learning Representation (ICLR), 2020

  7. Auditing Data Provenance in Text-Generation Models [pdf][code][slides]
    C.Song, V.Shmatikov
    In ACM SIGKDD Conference on Knowledge Discovery and Data Mining (KDD), 2019
    Oral Presentation

  8. Exploiting Unintended Feature Leakage in Collaborative Learning [pdf][code][talk][slides]
    L.Melis*, C.Song*, E. De Cristofaro, V.Shmatikov
    In IEEE Symposium on Security and Privacy (Oakland), 2019

  9. What Are Machine Learning Models Hiding? [pdf]
    V.Shamtikov, C.Song
    In Workshop on Hot Topics in Privacy Enhancing Technologies (HotPETs), 2018

  10. Kernel Distillation for Fast Gaussian Processes Prediction [pdf][code]
    C.Song*, Y.Sun*
    In NeurIPS Workshop on All of Bayesian Nonparametrics (BNP@NeurIPS), 2018
    Spotlight Presentation

  11. Predicting Clinical Outcomes from Large Scale Cancer Genomic Profiles with Deep Survival Models [pdf][code]
    S.Yousefi, F.Amrollahi, M.Amgad, C.Dong, J.E.Lewis, C.Song, D.A.Gutman, S.H.Halani, J.E.V.Vega, D.J.Brat, L.A.D.Cooper
    In Scientific Reports 7 (Nature), 2017

  12. Machine Learning Models that Remembers Too Much [pdf][code][talk][slides]
    C.Song, T.Risternpart, V.Shmatikov
    In ACM Conference on Computer and Communications Security (CCS), 2017

  13. Membership Inference Attacks Against Machine Learning Models [pdf][code][talk]
    R.Shokri, M.Stronati, C.Song, V.Shmatikov
    In IEEE Symposium on Security and Privacy (Oakland), 2017
    The Caspar Bowden Award for Outstanding Research in Privacy Enhancing Technologies 2018

  14. Learning Genomic Representations to Predict Clinical Outcomes in Cancer [pdf][code]
    S.Yousefi, C.Song, N.Nauata, L.Cooper
    In International Conference on Learning Representation Workshop (ICLRW), 2016


  1. Chiron: Privacy-preserving Machine Learning as a Service [pdf]
    T.Hunt, C.Song, R.Shokri, V.Shmatikov, E.Witchel
    In arXiv preprint, 2018

  2. Fooling OCR Systems with Adversarial Text Images [pdf][code by F.Tramèr et al]
    C.Song, V.Shmatikov
    In arXiv preprint, 2018