William Merrill

Assistant Professor, Toyota Technological Institute at Chicago

Postdoc, Allen Institute for Artificial Intelligence

Joined

July 2019

Names

William Merrill

Emails

****@gmail.com (Confirmed)

****@nyu.edu (Confirmed)

****@allenai.org (Confirmed)

Personal Links

Career & Education History

Assistant Professor

Toyota Technological Institute at Chicago (ttic.edu)

2026 – Present

Postdoc

Allen Institute for Artificial Intelligence (allenai.org)

2025 – 2026

Graduate student

Center for Data Science, New York University (nyu.edu)

2021 – 2025

Researcher

Allen Institute for Artificial Intelligence (allenai.org)

2019 – 2021

Undergrad student

Yale University (yale.edu)

2015 – 2019

Advisors, Relations & Conflicts

No relations added

Expertise

Natural Language Processing

Present

Formal Language and Automata

Present

Circuit Complexity

Present

Language Models

Present

Publications

OLMo: Accelerating the Science of Language Models
Dirk Groeneveld, Iz Beltagy, Evan Pete Walsh, Akshita Bhagia, Rodney Kinney, Oyvind Tafjord, Ananya Harsh Jha, Hamish Ivison, Ian Magnusson, Yizhong Wang, Shane Arora, Russell Authur, Khyathi Chandu, Arman Cohan, Jennifer Dumas, Yanai Elazar, Yuling Gu, Jack Hessel, Tushar Khot, William Merrill et al. (22 additional authors not shown)
- 20 Feb 2024
- OpenReview Archive Direct Upload
- Readers: Everyone
A Tale of Two Circuits: Grokking as Competition of Sparse and Dense Subnetworks
William Merrill, Nikolaos Tsilivis, Aman Shukla
- Published: 04 Mar 2023, Last Modified: 15 Feb 2026
- ME-FoMo 2023 Poster
- Readers: Everyone
ReCLIP: A Strong Zero-shot Baseline for Referring Expression Comprehension
Sanjay Subramanian, William Merrill, Trevor Darrell, Matt Gardner, Sameer Singh, Anna Rohrbach
- 16 Nov 2022
- OpenReview Archive Direct Upload
- Readers: Everyone
Transformers Implement First-Order Logic with Majority Quantifiers
William Merrill, Ashish Sabharwal
- 19 Oct 2022
- OpenReview Archive Direct Upload
- Readers: Everyone
Log-Precision Transformers are Uniform Threshold Circuits
William Merrill, Ashish Sabharwal
- 19 Oct 2022
- OpenReview Archive Direct Upload
- Readers: Everyone
Transformers Implement First-Order Logic with Majority Quantifiers
William Merrill, Ashish Sabharwal
- Published: 01 Feb 2023, Last Modified: 13 Feb 2023
- Submitted to ICLR 2023
- Readers: Everyone
Provable Limitations of Acquiring Meaning from Ungrounded Form: What Will Future Language Models Understand?
William Merrill, Yoav Goldberg, Roy Schwartz, Noah Smith
- 29 Oct 2021
- OpenReview Archive Direct Upload
- Readers: Everyone
Formal Language Theory Meets Modern NLP
William Merrill
- 2021 (modified: 11 Nov 2021)
- CoRR 2021
- Readers: Everyone
Effects of Parameter Norm Growth During Transformer Training: Inductive Bias from Gradient Descent
William Merrill, Vivek Ramanujan, Yoav Goldberg, Roy Schwartz, Noah A. Smith
- 2021 (modified: 11 Nov 2021)
- EMNLP (1) 2021
- Readers: Everyone
Competency Problems: On Finding and Removing Artifacts in Language Data
Matt Gardner, William Merrill, Jesse Dodge, Matthew E. Peters, Alexis Ross, Sameer Singh, Noah A. Smith
- 2021 (modified: 11 Nov 2021)
- EMNLP (1) 2021
- Readers: Everyone

View all 20 publications

Co-Authors

View all 94 co-authors