Sotabase
Home
Researchers
Career
·
Student
,
Stanford (Current)
2022–2027
·
Doctor of Philosophy - PhD, Computer Science
,
Stanford University
2022–2027
·
Bachelor of Science - BS, Computer Science
,
Carnegie Mellon University
2018–2022
Publications
(26)
One Step of Gradient Descent is Provably the Optimal In-Context Learner with One Layer of Linear Self-Attention
International Conference on Learning Representations · 2023
148
cited
Beyond NTK with Vanilla Gradient Descent: A Mean-Field Analysis of Neural Networks with Polynomial Width, Samples, and Time
Neural Information Processing Systems · 2023
16
cited
Formal Theorem Proving by Rewarding LLMs to Decompose Proofs Hierarchically
arXiv.org · 2024
12
cited
Optimal 𝓁1 Column Subset Selection and a Fast PTAS for Low Rank Approximation
ACM-SIAM Symposium on Discrete Algorithms · 2021
12
cited
Streaming and Distributed Algorithms for Robust Column Subset Selection
International Conference on Machine Learning · 2021
8
cited
A billiards-like dynamical system for attacking chess pieces
European journal of combinatorics (Print) · 2021
7
cited
Near-Linear Time and Fixed-Parameter Tractable Algorithms for Tensor Decompositions
Information Technology Convergence and Services · 2022
6
cited
Low Rank Approximation for General Tensor Networks
arXiv.org · 2022
1
cited
Treachery! When fairy chess pieces attack
2019
1
cited
BERT on Multitask Training: Bimodality, Ensemble, Round-robin, Text-encoding, and More
Comparing BERT Fine-Tuning Methods
Divide-and-Conquer CoT: RL for Reducing Latency via Parallel Reasoning
2026
Exploring Challenges in Multi-task BERT Optimization
Exploring Multi-Task Learning with Unbalanced Datasets and Gradient Surgery
ExtraBERT: Applying BERT to Multiple Downstream Language Tasks
Gradient Descent in Multi-Task Learning
Improving the performance of miniBERT and BERTaaR (BERT as a Recruiter)
Leverage Augmented Large Language Models to build Hyper Personalized Recommendation Systems
Linear and Kernel Classification in the Streaming Model: Improved Bounds for Heavy Hitters
Neural Information Processing Systems · 2021
Mastering minBERT: A True Balancing Act
Show all 26 papers →
Sotabase
Arvind Mahankali | Researcher Profile | Sotabase | Sotabase