Welcome to Le Fang's Homepage!

The domain "fangleai.com" (/ˈfaŋgəl eɪ ˈaɪ/) is interpreted as fangle (newly fashioned techniques, coincidently also my name) in artificial intelligence.

I am an engineer and researcher working on deep learning etc. Currently, I am working in Facebook Inc at bay area.

My CV can be found here.

Email: lefang at fb.com / lefang at buffalo.edu

Linkedin Github Google Scholar


I started my industrial career in July, 2020 as Research Scientist in Facebook Inc.

I obtained my PhD degree in June, 2020 in Department of Computer Science & Engineering, University at Buffalo-SUNY. My research interest is machine learning, especially deep learning & bayesian learning, with applications in natural language processing. My PhD advisors are Dr. Changyou Chen and Dr. Wen Dong. Prior to that, I have finished bachelor and master in applied physics in University of Science and Technology of China and University of Houston, respectively.


  • 07/08/2022 Our paper "Meta-Learning with Less Forgetting on Large-Scale Non-Stationary Task Distributions" gets accepted in ECCV 2022! [link]

  • 05/15/2022 Our paper "Improving Task-free Continual Learning by Distributionally Robust Memory Evolution" gets accepted in ICML 2022! [link]

  • 03/01/2022 Our paper "Learning to Learn and Remember Super Long Multi-Domain Task Sequence" gets accepted into CVPR 2022 (oral)! [link]

  • 07/22/2021 Our paper "Meta Learning on a Sequence of Imbalanced Domains with Difficulty Awareness" gets accepted into ICCV 2021! [link].

  • 01/04/2021 Our paper "Transformer-based Conditional Variational Autoencoder for Controllable Story Generation" is released in [arxiv].

  • 01/04/2021 Our paper "Outline to Story: Fine-grained Controllable Story Generation from Cascaded Events" is released in [arxiv].

  • 08/26/2020 I join the ads ranking core ML team in Facebook Inc at Sunnyvale, CA.

  • 06/22/2020 I start my first day as Research Scientist in Facebook Inc at Menlo Park, CA.

  • 06/01/2020 Finally, I obtained the PhD degree with dissertation "Towards Effective and Controllable Neural Text Generation". [link]

  • 08/23/2019 I get a semester-long research intern position about conditional text generation from JD digits AI lab at Silicon Valley, CA.

  • 08/16/2019 I have finished my summer intern in Seattle ads ranking team of Facebook. Thanks to my mentors and friends!

  • 08/13/2019 Our paper "Implicit Deep Latent Variable Models for Text Generation" gets accepted into EMNLP 2019! [arxiv] [Blog] [Poster] [Github]

  • 09/01/2018 This website is released.