Get started
Author

Eshaan Nichani

Observed name: Nichani, Eshaan
OpenAlex ID: A5004602170
Fine-Tuning Language Models with Just Forward Passes
2023 · Sadhika Malladi, Tianyu Gao, et al. · neural information processing systems · 79
PDF
Assessment of circulating copy number variant detection for cancer screening
2017 · Bhuvan Molparia, Eshaan Nichani, et al. · PLoS ONE · 42
PDF
Do Deeper Convolutional Networks Perform Better
2021 · Eshaan Nichani, Adityanarayanan Radhakrishnan, et al. · arXiv (Cornell University) · 9
Increasing Depth Leads to U-Shaped Test Risk in Over-parameterized Convolutional Networks
2020 · Nichani, Eshaan, Radhakrishnan, Adityanarayanan, et al. · arXiv (Cornell University) · 7
PDF
Causal Structure Discovery between Clusters of Nodes Induced by Latent Factors
2022 · Squires, Chandler, Yun, Annie, et al. · arXiv (Cornell University) · 6
PDF
How Transformers Learn Causal Structure with Gradient Descent
2024 · Nichani, Eshaan, Damian, Alex, et al. · arXiv (Cornell University) · 4
PDF
Metastable mixing of Markov chains: Efficiently sampling low temperature exponential random graphs
2024 · Guy Bresler, Dheeraj Nagaraj, et al. · The Annals of Applied Probability · 4
PDF
Provable Guarantees for Nonlinear Feature Learning in Three-Layer Neural Networks
2023 · Eshaan Nichani, Alex Damian, et al. · neural information processing systems · 4
PDF
Self-Stabilization: The Implicit Bias of Gradient Descent at the Edge of Stability
2022 · Damian, Alex, Nichani, Eshaan, et al. · arXiv (Cornell University) · 3
PDF
Smoothing the Landscape Boosts the Signal for SGD: Optimal Sample Complexity for Learning Single Index Models
2023 · Alex Damian, Eshaan Nichani, et al. · neural information processing systems · 2
PDF
Balancedness and Alignment are Unlikely in Linear Neural Networks.
2020 · Adityanarayanan Radhakrishnan, Eshaan Nichani, et al. · 2
Learning Hierarchical Polynomials of Multiple Nonlinear Features with Three-Layer Networks
2024 · Fu, Hengyu, Wang, Zihao, et al. · arXiv (Cornell University) · 1
PDF
Learning Hierarchical Polynomials with Three-Layer Neural Networks
2023 · Wang, Zihao, Nichani, Eshaan, et al. · arXiv (Cornell University) · 1
PDF
Identifying Good Directions to Escape the NTK Regime and Efficiently Learn Low-Degree Plus Sparse Polynomials
2022 · Eshaan Nichani, Yu Bai, et al. · neural information processing systems · 1
PDF
On Alignment in Deep Linear Neural Networks
2020 · Radhakrishnan, Adityanarayanan, Nichani, Eshaan, et al. · arXiv (Cornell University) · 1
PDF
Fine-Tuning Dynamics of In-Context Factual Recall in Transformers
2026 · Huang, Ruomin, Nichani, Eshaan, et al. · arXiv (Cornell University)
PDF
Sharp Capacity Thresholds in Linear Associative Memory: From Top-1 Retrieval to Tail-Average Learning
2026 · Barnfield, Nicholas, Kim, Juno, et al. · arXiv (Cornell University)
PDF
Sharp Capacity Scaling of Spectral Optimizers in Learning Associative Memory
2026 · Kim, Juno, Nichani, Eshaan, et al. · arXiv (Cornell University)
PDF
Quantitative Bounds for Length Generalization in Transformers
2025 · Izzo, Zachary, Nichani, Eshaan, et al. · arXiv (Cornell University)
PDF
On the Statistical Query Complexity of Learning Semiautomata: a Random Walk Approach
2025 · Giapitzakis, George, Fountoulakis, Kimon, et al. · arXiv (Cornell University)
PDF
Learning Compositional Functions with Transformers from Easy-to-Hard Data
2025 · Wang, Zixuan, Nichani, Eshaan, et al. · arXiv (Cornell University)
PDF
Emergence and scaling laws in SGD learning of shallow neural networks
2025 · Yunwei Ren, Eshaan Nichani, et al. · neural information processing systems
PDF
Understanding Factual Recall in Transformers via Associative Memories
2024 · Nichani, Eshaan, Lee, Jason D., et al. · arXiv (Cornell University)
PDF
Works count: 24
Citation count: 166
H-index: 5
I10-index: 2
Year
Topic
Type