Mike Lewis
Research Scientist, Facebook AI Research
Joined
October 2017
Names
Mike Lewis
Emails
****@fb.com (Confirmed)
Personal Links
Career & Education History
Research Scientist
Facebook AI Research (fb.com)
2016 – Present
PhD student
University of Edinburgh (ed.ac.uk)
2014 – 2016
Postdoc
University of Washington (washington.edu)
2010 – 2014
Advisors, Relations & Conflicts
Expertise
parsing
Present
dialog
, dialogue
Present
language models
Present
pretraining for NLP
Present
transformers
Present
Publications
An Empirical Study on Noisy Data and LLM Pretraining Loss Divergence
- ICLR 2026 Workshop DATA-FM
- Readers: Everyone
Latent Speech-Text Transformer
Yen-Ju Lu, Yashesh Gaur, Wei Zhou, Benjamin Muller, Jesus Villalba, Najim Dehak, Luke Zettlemoyer, Gargi Ghosh, Mike Lewis, Srini Iyer, Duc Le- ICLR 2026 Oral
- Readers: Everyone
FlexOLMo: Open Language Models for Flexible Data Use
Weijia Shi, Akshita Bhagia, Kevin Farhat, Niklas Muennighoff, Jacob Morrison, Evan Pete Walsh, Dustin Schwenk, Shayne Longpre, Jake Poznanski, Allyson Ettinger, Daogao Liu, Margaret Li, Mike Lewis, Wen-tau Yih, Dirk Groeneveld, Luca Soldaini, Kyle Lo, Noah A. Smith, Luke Zettlemoyer, Pang Wei Koh et al. (3 additional authors not shown)- NeurIPS 2025 spotlight
- Readers: Everyone
Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models
Weixin Liang, LILI YU, Liang Luo, Srini Iyer, Ning Dong, Chunting Zhou, Gargi Ghosh, Mike Lewis, Luke Zettlemoyer, Xi Victoria Lin- MCDC @ ICLR 2025
- Readers: Everyone
Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models
Weixin Liang, LILI YU, Liang Luo, Srini Iyer, Ning Dong, Chunting Zhou, Gargi Ghosh, Mike Lewis, Wen-tau Yih, Luke Zettlemoyer, Xi Victoria Lin- ICLR 2025 DeLTa Workshop Poster
- Readers: Everyone
Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models
Weixin Liang, LILI YU, Liang Luo, Srini Iyer, Ning Dong, Chunting Zhou, Gargi Ghosh, Mike Lewis, Wen-tau Yih, Luke Zettlemoyer, Xi Victoria Lin- ICLR 2025 Workshop World Models
- Readers: Everyone
Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models
Weixin Liang, LILI YU, Liang Luo, Srini Iyer, Ning Dong, Chunting Zhou, Gargi Ghosh, Mike Lewis, Wen-tau Yih, Luke Zettlemoyer, Xi Victoria Lin- Accepted by TMLR
- Readers: Everyone
Mixture-of-Transformers: A Sparse and Scalable Architecture for Multi-Modal Foundation Models
Weixin Liang, LILI YU, Liang Luo, Srini Iyer, Ning Dong, Chunting Zhou, Gargi Ghosh, Mike Lewis, Wen-tau Yih, Luke Zettlemoyer, Xi Victoria Lin- CPAL 2025 (Recent Spotlight Track)
- Readers: Everyone
MoMa: Efficient Early-Fusion Pre-training with Mixture of Modality-Aware Experts
Xi Victoria Lin, Akshat Shrivastava, Liang Luo, Srini Iyer, Mike Lewis, Gargi Ghosh, Luke Zettlemoyer, Armen Aghajanyan- ICLR 2025 Conference Withdrawn Submission
- Readers: Everyone
Co-Authors
- Abdelrahman Mohamed
- Adam Lerer
- Adi Renduchintala
- Akshat Shrivastava
- Akshita Bhagia
- Alane Suhr
- Aleksandra Piktus
- Alex Wang
- Alexander H Miller
- Alexander Wei
- Alexey Kozlov
- Ali Farhadi
- Allyson Ettinger
- Amjad Almahairi
- Anastasia Razdaibiedina
- Angela Fan
- Ankush Garg
- Anton Bakhtin
- Anuj Kumar
- Arash Einolghozati
- Ari Holtzman
- Armen Aghajanyan
- Artidoro Pagnoni
- Aston Zhang
- Athul Paul Jacob