Publications

• Contextual Speech Recognition with Difficult Negative Training Examples
  • Uri Alon, Golan Pundak, Tara N. Sainath
  • Research was done during an internship at Google NYC.
  • Accepted to ICASSP’2019: International Conference on Acoustics, Speech, and Signal Processing
  • [PDF][Poster][BibTex]
• code2seq: Generating Sequences from Structured Representations of Code
  • Uri Alon, Shaked Brody, Omer Levy, Eran Yahav
  • Accepted to ICLR’2019: International Conference on Learning Representations
  • Online demo: https://code2seq.org
  • [PDF][Poster][Code and trained model][BibTeX]
  • Datasets: [Java-small][Java-med][Java-large]
• code2vec: Learning Distributed Representations of Code
  • Uri Alon, Meital Zilberstein, Omer Levy, Eran Yahav
  • Appeared at POPL’2019: Principles of Programming Languages
  • Online demo: https://www.code2vec.org
  • [PDF][Slides] [Video][BibTeX]
  • [Code and trained model]
  • [Token vectors][Method name vectors]
• A General Path-Based Representation for Predicting Program Properties
  • Uri Alon, Meital Zilberstein, Omer Levy, Eran Yahav
  • Appeared at PLDI’2018: Programming Languages Design and Implementation
  • [PDF][Slides][Video][BibTeX]
  • [JS Code][JS Dataset][Python Dataset]
  • [JS embeddings]

Technical Reports

• Lingvo: a Modular and Scalable Framework for Sequence-to-Sequence Modeling
  • Participated in development during an internship at Google
  • [PDF]