Sotabase

Career

· Founder Contributing Writer, SAIL Media (Current)2025–
· Research Scientist, Allen Institute for AI (AI2)2023–
· Senior Research Scientist, Allen Institute for AI (AI2)2023–
· Research Scientist RLHF Team Lead, Hugging Face2022–2023
· Founder, Interconnects AI2022–
· Research Scientist Intern, DeepMind2021–2021
· Instructor, University of California, Berkeley2020–2020
· Research Intern Student Researcher, Facebook AI2019–2020
· PHD Candidate, University of California, Berkeley2017–2022
· Undergraduate Researcher, SonicMEMs Laboratory, Cornell University2016–2017
· Intern, Tesla Motors2015–2015

Publications (49)

OLMo: Accelerating the Science of Language Models
Annual Meeting of the Association for Computational Linguistics · 2024
563
cited
Zephyr: Direct Distillation of LM Alignment
arXiv.org · 2023
532
cited
TÜLU 3: Pushing Frontiers in Open Language Model Post-Training
arXiv.org · 2024
479
cited
Dolma: an Open Corpus of Three Trillion Tokens for Language Model Pretraining Research
Annual Meeting of the Association for Computational Linguistics · 2024
399
cited
Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Multimodal Models
arXiv.org · 2024
356
cited
RewardBench: Evaluating Reward Models for Language Modeling
arXiv.org · 2024
352
cited
WildGuard: Open One-Stop Moderation Tools for Safety Risks, Jailbreaks, and Refusals of LLMs
Neural Information Processing Systems · 2024
249
cited
Camels in a Changing Climate: Enhancing LM Adaptation with Tulu 2
arXiv.org · 2023
238
cited
A Survey on Data Selection for Language Models
Trans. Mach. Learn. Res. · 2024
215
cited
OLMoE: Open Mixture-of-Experts Language Models
arXiv.org · 2024
164
cited
Low-Level Control of a Quadrotor With Deep Model-Based Reinforcement Learning
IEEE Robotics and Automation Letters · 2019
158
cited
2 OLMo 2 Furious
arXiv.org · 2024
150
cited
Spurious Rewards: Rethinking Training Signals in RLVR
arXiv.org · 2025
118
cited
Objective Mismatch in Model-based Reinforcement Learning
Conference on Learning for Dynamics & Control · 2020
108
cited
On the Importance of Hyperparameter Optimization for Model-based Reinforcement Learning
International Conference on Artificial Intelligence and Statistics · 2021
108
cited
Molmo and PixMo: Open Weights and Open Data for State-of-the-Art Vision-Language Models
Computer Vision and Pattern Recognition · 2024
105
cited
Unpacking DPO and PPO: Disentangling Best Practices for Learning from Preference Feedback
Neural Information Processing Systems · 2024
98
cited
Toward Controlled Flight of the Ionocraft: A Flying Microrobot Using Electrohydrodynamic Thrust With Onboard Sensing and No Moving Parts
IEEE Robotics and Automation Letters · 2018
65
cited
MBRL-Lib: A Modular Library for Model-based Reinforcement Learning
arXiv.org · 2021
52
cited
The History and Risks of Reinforcement Learning and Human Feedback
2023
49
cited
Sotabase
Nathan Lambert | Researcher Profile | Sotabase | Sotabase