FlexOlmo: Open Language Models for Flexible Data Use

Akshita Bhagia's homepage says she is a Senior Research Engineer on the AllenNLP team at Ai2, involved in R&D for NLP and in building and evaluating large language models.

Luke Zettlemoyer works on empirical methods for natural language semantics, machine learning, new tasks and datasets, and self-supervision for pre-training.

Hannaneh Hajishirzi is Senior Director at Ai2 and a professor of computer science at the University of Washington. Her public materials focus on generative AI, reasoning, open language models, pre-training, and evaluation.

The Allen School lists Pang Wei Koh as an assistant professor at the University of Washington and a visiting research scientist at the Allen Institute for AI. The same page lists expertise in explainable AI, generative AI, machine learning, and natural language processing.

Daogao Liu's homepage says he is currently a visiting postdoctoral researcher at Google. It says his research focuses on machine learning and theoretical computer science, especially differential privacy, stochastic optimization, and algorithm design.

Noah A. Smith is the University of Washington's inaugural Vice Provost for Artificial Intelligence and a professor in the Paul G. Allen School of Computer Science & Engineering. He is also Senior Director of NLP Research at the Allen Institute for AI.

Sewon Min works on natural language processing and machine learning, with a focus on large language models. She earned a Ph.D. in Computer Science & Engineering from the University of Washington and a B.S. in Computer Science and Engineering from Seoul National University.

Luca Soldaini's public materials describe him as a Senior Applied Research Scientist at the Allen Institute for AI working on data-centric NLP, information retrieval, and the OLMo language model effort.

Jacob Morrison's homepage says he is a PhD student at UW and a student researcher on the AllenNLP team at Ai2, where he mainly works on language modeling.

Allyson Ettinger is a senior research scientist at Ai2 working on the OLMo pre-training team. Her public materials focus on natural language understanding, language model behavior, and learning from text and experience.

Scott Wen-tau Yih is a research scientist at Meta FAIR whose recent work focuses on neural retrieval and retrieval-augmented generation; his homepage lists prior roles at Ai2 and Microsoft Research.

Weijia Shi is a Ph.D. candidate at the University of Washington. Her research focuses on building augmented and modular architectures to improve the quality, efficiency, and capabilities of language models.

Niklas Muennighoff researches language models and evaluation at Stanford and has publicly led open-model work including OLMoE.

Kyle Lo's public homepage says he formerly served as a research scientist at the Allen Institute for AI, where he co-led OLMo. It describes current work on large-scale pretraining, data mixing, benchmark validity, and research-supporting data and software artifacts.

Jakub "Jake" Poznanski identifies himself on his personal site as a Carnegie Mellon computer science graduate and describes leading an OCR toolkit on GitHub.

Kevin Farhat's homepage says he is a research engineer at Ai2 working on language models, distributed AI, and privacy-preserving training. Public sources also list him as a coauthor of FlexOlmo.

Shayne Longpre's public profiles identify him as a PhD candidate at MIT focused on data-centric AI, language models, and their societal impact.

OpenReview lists Evan Pete Walsh as a researcher at the Allen Institute for Artificial Intelligence, and recent public reports credit Pete Walsh on OLMo Hybrid and related open-model work.

Public sources reviewed here support Ali Farhadi's University of Washington faculty role and his OLMo 3 authorship. They do not provide a current official title for any other organization.

Public sources identify Dirk Groeneveld as an OLMo 3 coauthor; a 2022 USC ISI speaker bio described him as a principal engineer at AI2 and technical lead for the AllenNLP Platform team.

Canonical link

Bhagia, Akshita

Zettlemoyer, Luke

Hajishirzi, Hannaneh

Koh, Pang Wei

Liu, Daogao

Smith, Noah A.

Min, Sewon

Soldaini, Luca

Morrison, Jacob

Ettinger, Allyson

Yih, Wen-tau

Shi, Weijia

Muennighoff, Niklas

Lo, Kyle

Poznanski, Jake

Farhat, Kevin

Longpre, Shayne

Walsh, Pete

Farhadi, Ali

Groeneveld, Dirk