Jack Merullo

34 papers A* 7B 1Journal 20Unranked 6
YearRankTypeTitle / Venue / Authors
2026 J jnl
Features as Rewards: Scalable Supervision for Open-Ended Tasks via Interpretability.
CoRR
Aaditya Vikram Prasad, Connor Watts, Jack Merullo, Dhruvil Gala, Owen Lewis, Thomas McGrath, Ekdeep Singh Lubana
2026 J jnl
Reasoning Theater: Disentangling Model Beliefs from Chain-of-Thought.
CoRR
Siddharth Boppana, Annabel Ma, Max Loeffler, Raphaël Sarfati, Eric Bigelow, Atticus Geiger, Owen Lewis, Jack Merullo
2026 J jnl
The Shape of Beliefs: Geometry, Dynamics, and Interventions along Representation Manifolds of Language Models' Posteriors.
CoRR
Raphaël Sarfati, Eric J. Bigelow, Daniel Wurgaft, Jack Merullo, Atticus Geiger, Owen Lewis, Tom McGrath, Ekdeep Singh Lubana
2025 A* conf
Dual Process Learning: Controlling Use of In-Context vs. In-Weights Strategies with Weight Forgetting.
ICLR
Suraj Anand, Michael A. Lepori, Jack Merullo, Ellie Pavlick
2025 J jnl
From Memorization to Reasoning in the Spectrum of Loss Curvature.
CoRR
Jack Merullo, Srihita Vatsavaya, Lucius Bushnaq, Owen Lewis
2025 J jnl
I Have No Mouth, and I Must Rhyme: Uncovering Internal Phonetic Representations in LLaMA 3.2.
CoRR
Jack Merullo, Arjun Khurana, Oliver McLaughlin
2025 A* conf
On Linear Representations and Pretraining Data Frequency in Language Models.
ICLR
Jack Merullo, Noah A. Smith, Sarah Wiegreffe, Yanai Elazar
2025 J jnl
On Linear Representations and Pretraining Data Frequency in Language Models.
CoRR
Jack Merullo, Noah A. Smith, Sarah Wiegreffe, Yanai Elazar
2025 J jnl
Transferring Features Across Language Models With Model Stitching.
CoRR
Alan Chen, Jack Merullo, Alessandro Stolfo, Ellie Pavlick
2024 J jnl
$100K or 100 Days: Trade-offs when Pre-Training with Academic Resources.
CoRR
Apoorv Khandelwal, Tian Yun, Nihal V. Nayak, Jack Merullo, Stephen H. Bach, Chen Sun, Ellie Pavlick
2024 A* conf
Axiomatic Causal Interventions for Reverse Engineering Relevance Computation in Neural Retrieval Models.
SIGIR
Catherine Chen, Jack Merullo, Carsten Eickhoff
2024 J jnl
Axiomatic Causal Interventions for Reverse Engineering Relevance Computation in Neural Retrieval Models.
CoRR
Catherine Chen, Jack Merullo, Carsten Eickhoff
2024 A* conf
Circuit Component Reuse Across Tasks in Transformer Language Models.
ICLR
Jack Merullo, Carsten Eickhoff, Ellie Pavlick
2024 — conf
Does CLIP Bind Concepts? Probing Compositionality in Large Image Models.
EACL (Findings)
Martha Lewis, Nihal V. Nayak, Peilin Yu, Jack Merullo, Qinan Yu, Stephen H. Bach, Ellie Pavlick
2024 J jnl
Dual Process Learning: Controlling Use of In-Context vs. In-Weights Strategies with Weight Forgetting.
CoRR
Suraj Anand, Michael A. Lepori, Jack Merullo, Ellie Pavlick
2024 — conf
Language Models Implement Simple Word2Vec-style Vector Arithmetic.
NAACL-HLT
Jack Merullo, Carsten Eickhoff, Ellie Pavlick
2024 A* conf
Talking Heads: Understanding Inter-Layer Communication in Transformer Language Models.
NeurIPS
Jack Merullo, Carsten Eickhoff, Ellie Pavlick
2024 J jnl
Talking Heads: Understanding Inter-layer Communication in Transformer Language Models.
CoRR
Jack Merullo, Carsten Eickhoff, Ellie Pavlick
2024 B conf
Transformer Mechanisms Mimic Frontostriatal Gating Operations When Trained on Human Working Memory Tasks.
CogSci
Aaron Traylor, Jack Merullo, Michael J. Frank, Ellie Pavlick
2024 J jnl
Transformer Mechanisms Mimic Frontostriatal Gating Operations When Trained on Human Working Memory Tasks.
CoRR
Aaron Traylor, Jack Merullo, Michael J. Frank, Ellie Pavlick
2023 A* conf
Characterizing Mechanisms for Factual Recall in Language Models.
EMNLP
Qinan Yu, Jack Merullo, Ellie Pavlick
2023 J jnl
Characterizing Mechanisms for Factual Recall in Language Models.
CoRR
Qinan Yu, Jack Merullo, Ellie Pavlick
2023 J jnl
Circuit Component Reuse Across Tasks in Transformer Language Models.
CoRR
Jack Merullo, Carsten Eickhoff, Ellie Pavlick
2023 J jnl
Language Models Implement Simple Word2Vec-style Vector Arithmetic.
CoRR
Jack Merullo, Carsten Eickhoff, Ellie Pavlick
2023 A* conf
Linearly Mapping from Image to Text Space.
ICLR
Jack Merullo, Louis Castricato, Carsten Eickhoff, Ellie Pavlick
2023 — conf
ezCoref: Towards Unifying Annotation Guidelines for Coreference Resolution.
EACL (Findings)
Ankita Gupta, Marzena Karpinska, Wenlong Zhao, Kalpesh Krishna, Jack Merullo, Luke Yeh, Mohit Iyyer, Brendan O'Connor
2022 — conf
ACQuA: Anomaly Classification with Quasi-Attractors.
CinC
William Rudman, Jack Merullo, Laura Mercurio, Carsten Eickhoff
2022 J jnl
Does CLIP Bind Concepts? Probing Compositionality in Large Image Models.
CoRR
Martha Lewis, Qinan Yu, Jack Merullo, Ellie Pavlick
2022 J jnl
Linearly Mapping from Image to Text Space.
CoRR
Jack Merullo, Louis Castricato, Carsten Eickhoff, Ellie Pavlick
2022 — conf
Pretraining on Interactions for Learning Grounded Affordance Representations.
*SEM@NAACL-HLT
Jack Merullo, Dylan Ebert, Carsten Eickhoff, Ellie Pavlick
2022 J jnl
Pretraining on Interactions for Learning Grounded Affordance Representations.
CoRR
Jack Merullo, Dylan Ebert, Carsten Eickhoff, Ellie Pavlick
2022 J jnl
ezCoref: Towards Unifying Annotation Guidelines for Coreference Resolution.
CoRR
Ankita Gupta, Marzena Karpinska, Wenlong Zhao, Kalpesh Krishna, Jack Merullo, Luke Yeh, Mohit Iyyer, Brendan T. O'Connor
2019 — conf
Investigating Sports Commentator Bias within a Large Corpus of American Football Broadcasts.
EMNLP/IJCNLP (1)
Jack Merullo, Luke Yeh, Abram Handler, Alvin Grissom II, Brendan O'Connor, Mohit Iyyer
2019 J jnl
Investigating Sports Commentator Bias within a Large Corpus of American Football Broadcasts.
CoRR
Jack Merullo, Luke Yeh, Abram Handler, Alvin Grissom II, Brendan O'Connor, Mohit Iyyer