Follow
Harshay Shah
Harshay Shah
Verified email at mit.edu - Homepage
Title
Cited by
Cited by
Year
The Pitfalls of Simplicity Bias in Neural Networks
H Shah, K Tamuly, A Raghunathan, P Jain, P Netrapalli
Neural Information Processing Systems (NeurIPS), 2020, 2020
4032020
Do Input Gradients Highlight Discriminative Features?
H Shah, P Jain, P Netrapalli
Neural Information Processing Systems (NeurIPS), 2021, 2021
712021
ModelDiff: A Framework for Comparing Learning Algorithms
H Shah, SM Park, A Ilyas, A Madry
International Conference on Machine Learning (ICML), 2023, 2022
282022
Decomposing and editing predictions by modeling model computation
H Shah, A Ilyas, A Madry
arXiv preprint arXiv:2404.11534, 2024
13*2024
ContextCite: Attributing Model Generation to Context
B Cohen-Wang, H Shah, K Georgiev, A Madry
arXiv preprint arXiv:2409.00729, 2024
82024
Growing Attributed Networks through Local Processes
H Shah, S Kumar, H Sundaram
The World Wide Web Conference (WWW), 2019, 2019
3*2019
Parameters vs FLOPs: Scaling Laws for Optimal Sparsity for Mixture-of-Experts Language Models
S Abnar, H Shah, D Busbridge, AME Ali, J Susskind, V Thilak
arXiv preprint arXiv:2501.12370, 2025
2025
Number of Connected Components in a Graph: Estimation via Counting Patterns
A Khetan, H Shah, S Oh
arXiv preprint arXiv:1812.00139, 2018
2018
The system can't perform the operation now. Try again later.
Articles 1–8