Naomi Saphra

Most-cited papers

DyNet: The Dynamic Neural Network Toolkit
Graham Neubig, Chris Dyer, Yoav Goldberg, A. Matthews +21 more
2017arXiv.org399 citations
Semantic Scholar
A taxonomy and review of generalization research in NLP
Dieuwke Hupkes, Mario Giulianelli, Verna Dankers, Mikel Artetxe +16 more
2022Nature Machine Intelligence164 citations Open access
Semantic Scholar
Sudden Drops in the Loss: Syntax Acquisition, Phase Transitions, and Simplicity Bias in MLMs
Angelica Chen, Ravid Shwartz-Ziv, Kyunghyun Cho, Matthew L. Leavitt +1 more
2023International Conference on Learning Representations143 citations Open access
Semantic Scholar
Edinburgh Research Explorer Understanding Learning Dynamics Of Language Models with SVCCA
Naomi Saphra
124 citations
Semantic Scholar
Understanding Objects in Detail with Fine-Grained Attributes
A. Vedaldi, S. Mahendran, Stavros Tsogkas, Subhransu Maji +10 more
20142014 IEEE Conference on Computer Vision and Pattern Recognition119 citations Open access
Semantic Scholar
The MultiBERTs: BERT Reproductions for Robustness Analysis
T. Sellam, Steve Yadlowsky, Ian Tenney, Jason Wei +8 more
2021International Conference on Learning Representations116 citations
Semantic Scholar
Understanding Privacy-Related Questions on Stack Overflow
Mohammad Tahaei, Kami Vaniea, Naomi Saphra
2020International Conference on Human Factors in Computing Systems115 citations Open access
Semantic Scholar
Pareto Probing: Trading-Off Accuracy and Complexity
Tiago Pimentel, Naomi Saphra, Adina Williams, Ryan Cotterell
2020Conference on Empirical Methods in Natural Language Processing75 citations Open access
Semantic Scholar
Linear Connectivity Reveals Generalization Strategies
Jeevesh Juneja, Rachit Bansal, Kyunghyun Cho, João Sedoc +1 more
2022International Conference on Learning Representations61 citations Open access
Semantic Scholar
An Algerian Arabic-French Code-Switched Corpus
Ryan Cotterell, Adithya Renduchintala, Naomi Saphra, Chris Callison-Burch
201452 citations
Semantic Scholar
A Non-Linear Structural Probe
Jennifer C. White, Tiago Pimentel, Naomi Saphra, Ryan Cotterell
2021North American Chapter of the Association for Computational Linguistics44 citations Open access
Semantic Scholar
PolyPythias: Stability and Outliers across Fifty Language Model Pre-Training Runs
Oskar van der Wal, Pietro Lesci, Max Müller-Eberstein, Naomi Saphra +3 more
2025International Conference on Learning Representations39 citations
Semantic Scholar
Recite, Reconstruct, Recollect: Memorization in LMs as a Multifaceted Phenomenon
USVSN Sai Prashanth, Alvin Deng, Kyle O'Brien, V JyothirS +8 more
2024International Conference on Learning Representations36 citations
Semantic Scholar
A Framework for (Under)specifying Dependency Syntax without Overloading Annotators
Nathan Schneider, Brendan T. O'Connor, Naomi Saphra, David Bamman +4 more
2013LAW@ACL26 citations
Semantic Scholar
Dynamic Masking Rate Schedules for MLM Pretraining
Zachary Ankner, Naomi Saphra, Davis W. Blalock, Jonathan Frankle +1 more
2023Conference of the European Chapter of the Association for Computational Linguistics22 citations Open access
Semantic Scholar
Latent State Models of Training Dynamics
Michael Y Hu, Angelica Chen, Naomi Saphra, Kyunghyun Cho
2023Trans. Mach. Learn. Res.21 citations Open access
Semantic Scholar
LSTMS Compose — and Learn — Bottom-Up
Naomi Saphra, A. Lopez
2020Findings18 citations Open access
Semantic Scholar
Benchmarking Compositionality with Formal Languages
Josef Valvoda, Naomi Saphra, Jonathan Rawski, Adina Williams +1 more
2022International Conference on Computational Linguistics16 citations
Semantic Scholar
Can Interpretation Predict Behavior on Unseen Data?
Victoria R. Li, Jenny Kaufmann, Tian Qin, Martin Wattenberg +2 more
2025arXiv.org13 citations
Semantic Scholar
Distributional Scaling Laws for Emergent Capabilities
Rosie Zhao, Tian Qin, David Alvarez-Melis, S. Kakade +1 more
2025arXiv.org12 citations
Semantic Scholar
Language Models Learn POS First
Naomi Saphra, A. Lopez
2018BlackboxNLP@EMNLP12 citations Open access
Semantic Scholar
Do Activation Verbalization Methods Convey Privileged Information?
Millicent Li, Alberto Mario Ceballos Arroyo, Giordano Rogers, Naomi Saphra +1 more
202511 citations
Semantic Scholar
Sometimes I am a Tree: Data Drives Unstable Hierarchical Generalization
Tian Qin, Naomi Saphra, David Alvarez-Melis
2024arXiv.org11 citations
Semantic Scholar
AMRICA: an AMR Inspector for Cross-language Alignments
Naomi Saphra, A. Lopez
2015North American Chapter of the Association for Computational Linguistics11 citations Open access
Semantic Scholar
Training dynamics of neural language models
Naomi Saphra
202110 citations
Semantic Scholar
First Tragedy, then Parse: History Repeats Itself in the New Era of Large Language Models
Naomi Saphra, Eve Fleisig, Kyunghyun Cho, Adam Lopez
2023arXiv.org9 citations
Semantic Scholar
TRAM: Bridging Trust Regions and Sharpness Aware Minimization
Tom Sherborne, Naomi Saphra, Pradeep Dasigi, Hao Peng
2023International Conference on Learning Representations7 citations Open access
Semantic Scholar
Data Drives Unstable Hierarchical Generalization in LMs
Tian Qin, Naomi Saphra, David Alvarez-Melis
2025Conference on Empirical Methods in Natural Language Processing5 citations Open access
Semantic Scholar
Towards a Detailed Understanding of Objects and Scenes in Natural Images
Matthew B. Blaschko, Ross B. Girshick, Juho Kannala, I. Kokkinos +9 more
20125 citations
Semantic Scholar
Attribute Diversity Determines the Systematicity Gap in VQA
Ian Berlot-Attwell, K. Agrawal, Annabelle Michael Carrell, Yash Sharma +1 more
2023Conference on Empirical Methods in Natural Language Processing4 citations
Semantic Scholar
One Venue, Two Conferences: The Separation of Chinese and American Citation Networks
Bingchen Zhao, Yuling Gu, Jessica Zosa Forde, Naomi Saphra
2022arXiv.org3 citations Open access
Semantic Scholar
Word Interdependence Exposes How LSTMs Compose Representations
Naomi Saphra, A. Lopez
2020arXiv.org3 citations
Semantic Scholar
Edinburgh Research Explorer Evaluating Informal-Domain Word Representations With UrbanDictionary
Naomi Saphra, A. Lopez
3 citations
Semantic Scholar
Towards out-of-distribution generalization in large-scale astronomical surveys: robust networks learn similar representations
Yash Gondhalekar, Sultan Hassan, Naomi Saphra, S. Andrianomena
2023arXiv.org2 citations
Semantic Scholar
Sparsity Emerges Naturally in Neural Language Models
Naomi Saphra, A. Lopez
2019arXiv.org1 citations
Semantic Scholar
Shared Circuits for Shared Grammar: Tracing Subject-Verb Agreement Across Languages
Isabella Gidi, Antonio Almudévar, C. Park, Naomi Saphra +1 more
20260 citations
Semantic Scholar
Learning Transductions to Test Systematic Compositionality
Josef Valvoda, Naomi Saphra, Jonathan Rawski, Ryan Cotterell +1 more
2022arXiv.org0 citations Open access
Semantic Scholar
Automatic Verification of Boundary Annotations from Mechanical Turk
Naomi Saphra, Juho Kannala, Esa Rahtu, Matthew B. Blaschko
20120 citations
Semantic Scholar