Saibo Geng

Saibo Geng

PhD student in EPFL

EPFL Data Science Lab

Biography

Hey there! I’m currently a Ph.D. student at EPFL’s Data Science Lab, where I’m advised by Prof. Robert West. Prior to joining EPFL, I received my B.S. in Physics from University Paris-Saclay, France.

News

  • I started Awesome-LLM-Constrained-Decoding project to centralize resources on constrained decoding for LLM.
  • I will be joining Microsoft(Redmond) as a research intern in the summer of 2024.
  • I have built an open source library for grammar-constrained decoding called Transformers-CFG, feel free to check it out!

About Me

My research interests include:

  • Develop new constrained decoding research techniques for increasing LLM inference quality and/or efficiency.
  • Design principled evaluation frameworks and benchmarks for measuring the effects of constrained decoding on a model.

Please check out my Google Scholar for more details on my publications.

If you have any questions or would like to collaborate, feel free to reach out to me via email or connect with me on Twitter

Outside of research, I enjoy reading all kinds of books, traveling and learning different languages. Having spent many years in France and Switzerland, I am fluent in French, English, and native in Chinese. I am currently picking up German and Korean in my free time.

Interests
  • Grammar-constrained Decoding
  • Efficient Decoding methods
  • Domain-Specific Language generation with LLM
Education
  • PhD in Computer Science, 2022-Now

    EPFL

Updates

Grammar-Constrained Decoding: A New Paradigm for Reliable and Efficient Language Model Generation
Invited talk on Grammar-Constrained Decoding at NexThink (Switzerland)
Transformers-CFG is now available on GitHub
A grammar-constrained decoding library for LLM that supports CFG, EBNF, and unicode grammars
Low-memory Beam Search is merged into Huggingface Transformers
A low-memory beam search implementation that trades off speed for memory

Recent Posts