VS
Virginia Smith
cs.LGstat.MLcs.CLcs.CRcs.AIcs.DCmath.OCcs.ITmath.ITcs.CV
On Valency
published · living versionsW_u7fa5rax·v1 · currentpublished
MLSys: The New Frontier of Machine Learning Systems
with Alexander Ratner, Dan Alistarh, Gustavo Alonso, David G. Andersen +64
1 version
Preprints & journals
63 papers in the corpus · 2013–2026What Survives When You Compress a Recursive Reasoner for the Edge?2606.26488v2 · Pearse Jim, Steven Kolawole, Opegbemi Matthias Busoye et al.2026 · 0 citationsarXiv
EpiKV: Epiphany-Aware KV Cache Eviction Without the Attention Matrix2606.26472v4 · Steven Kolawole, Virginia Smith2026 · 0 citationsarXiv
SAEs Can Improve Unlearning: Dynamic Sparse Autoencoder Guardrails for Precision Unlearning in LLMs2504.08192v2 · Aashiq Muhamed, Jacopo Bonato, Mona Diab et al.2025 · 0 citationsarXiv
MOLE: Detecting Insider Threats in AI Agents2609.06966v1 · Aashiq Muhamed, Virginia Smith2026 · 0 citationsarXiv
Membership Inference Attacks for Unseen Classes2506.06488v3 · Pratiksha Thaker, Neil Kale, Zhiwei Steven Wu et al.2025 · 0 citationsarXiv
Everybody Prune Now: Structured Pruning of LLMs with only Forward Passes2402.05406v4 · Steven Kolawole, Lucio Dery, Jean-Francois Kagy et al.2024 · 6 citationsarXiv
Semantic Agreement Enables Efficient Open-Ended LLM Cascades2509.21837v3 · Duncan Soiffer, Steven Kolawole, Virginia Smith2025 · 0 citationsarXiv
RefusalBench: Generative Evaluation of Selective Refusal in Grounded Language Models2510.10390v1 · Aashiq Muhamed, Leonardo F. R. Ribeiro, Markus Dreyer et al.2025 · 0 citationsarXiv
Agreement-Based Cascading for Efficient Inference2407.02348v4 · Steven Kolawole, Don Dennis, Ameet Talwalkar et al.2024 · 0 citationsTMLR 2025
On the Benefits of Public Representations for Private Transfer Learning under Distribution Shift2312.15551v5 · Pratiksha Thaker, Amrith Setlur, Zhiwei Steven Wu et al.2023 · 0 citationsarXiv
e3: Learning to Explore Enables Extrapolation of Test-Time Compute for LLMs2506.09026v2 · Amrith Setlur, Matthew Y. R. Yang, Charlie Snell et al.2025 · 0 citationsarXiv
NeurIPS 2023 Competition: Privacy Preserving Federated Learning Document VQA2411.03730v2 · Marlon Tobaben, Mohamed Ali Souibgui, Ruben Tito et al.2024 · 0 citationsTransactions on Machine Learning Research, ISSN 2835-8856, 2025
Position: Mechanistic Interpretability Should Prioritize Feature Consistency in SAEs2505.20254v1 · Xiangchen Song, Aashiq Muhamed, Yujia Zheng et al.2025 · 0 citationsarXiv
Position: LLM Unlearning Benchmarks are Weak Measures of Progress2410.02879v2 · Pratiksha Thaker, Shengyuan Hu, Neil Kale et al.2024 · 6 citationsarXiv
Exact Unlearning of Finetuning Data via Model Merging at Scale2504.04626v1 · Kevin Kuo, Amrith Setlur, Kartik Srinivas et al.2025 · 0 citationsarXiv
CoRAG: Collaborative Retrieval-Augmented Generation2504.01883v1 · Aashiq Muhamed, Mona Diab, Virginia Smith2025 · 1 citationarXiv
Decoding Dark Matter: Specialized Sparse Autoencoders for Interpreting Rare Concepts in Foundation Models2411.00743v1 · Aashiq Muhamed, Mona Diab, Virginia Smith2024 · 1 citationarXiv
Grass: Compute Efficient Low-Memory LLM Training with Structured Sparse Gradients2406.17660v1 · Aashiq Muhamed, Oscar Li, David Woodruff et al.2024 · 3 citationsarXiv
RL on Incorrect Synthetic Data Scales the Efficiency of LLM Math Reasoning by Eight-Fold2406.14532v1 · Amrith Setlur, Saurabh Garg, Xinyang Geng et al.2024 · 6 citationsarXiv
Privacy Amplification for the Gaussian Mechanism via Bounded Support2403.05598v1 · Shengyuan Hu, Saeed Mahloujifar, Virginia Smith et al.2024 · 0 citationsarXiv
Many-Objective Multi-Solution Transport2403.04099v1 · Ziyue Li, Tian Li, Virginia Smith et al.2024 · 0 citationsarXiv
Progressive Ensemble Distillation: Building Ensembles for Efficient Inference2302.10093v2 · Don Kurian Dennis, Abhishek Shetty, Anish Sevekari et al.2023 · 3 citationsarXiv
Private Multi-Task Learning: Formulation and Applications to Federated Learning2108.12978v3 · Shengyuan Hu, Zhiwei Steven Wu, Virginia Smith2021 · 4 citationsarXiv
Bitrate-Constrained DRO: Beyond Worst Case Robustness To Unknown Group Shifts2302.02931v2 · Amrith Setlur, Don Dennis, Benjamin Eysenbach et al.2023 · 1 citationICLR 2023
Differentially Private Adaptive Optimization with Delayed Preconditioners2212.00309v2 · Tian Li, Manzil Zaheer, Ken Ziyu Liu et al.2022 · 3 citationsarXiv
On Tilted Losses in Machine Learning: Theory and Applications2109.06141v3 · Tian Li, Ahmad Beirami, Maziar Sanjabi et al.2021 · 9 citationsarXiv
On Noisy Evaluation in Federated Hyperparameter Tuning2212.08930v4 · Kevin Kuo, Pratiksha Thaker, Mikhail Khodak et al.2022 · 0 citationsarXiv
Validating Large Language Models with ReLM2211.15458v2 · Michael Kuchnik, Virginia Smith, George Amvrosiadis2022 · 8 citationsarXiv
Federated Learning as a Network Effects Game2302.08533v1 · Shengyuan Hu, Dung Daniel Ngo, Shuran Zheng et al.2023 · 0 citationsarXiv
Maximizing Global Model Appeal in Federated Learning2205.14840v2 · Yae Jee Cho, Divyansh Jhunjhunwala, Tian Li et al.2022 · 3 citationsarXiv
Fair Federated Learning via Bounded Group Loss2203.10190v3 · Shengyuan Hu, Zhiwei Steven Wu, Virginia Smith2022 · 9 citationsarXiv
Motley: Benchmarking Heterogeneity and Personalization in Federated Learning2206.09262v6 · Shanshan Wu, Tian Li, Zachary Charles et al.2022 · 13 citationsarXiv
Private Adaptive Optimization with Side Information2202.05963v2 · Tian Li, Manzil Zaheer, Sashank J. Reddi et al.2022 · 3 citationsarXiv
Label Leakage and Protection in Two-party Split Learning2102.08504v3 · Oscar Li, Jiankai Sun, Xin Yang et al.2021 · 63 citationsarXiv
Plumber: Diagnosing and Removing Performance Bottlenecks in Machine Learning Data Pipelines2111.04131v2 · Michael Kuchnik, Ana Klimovic, Jiri Simsa et al.2021 · 11 citationsarXiv
Federated Hyperparameter Tuning: Challenges, Baselines, and Connections to Weight-Sharing2106.04502v2 · Mikhail Khodak, Renbo Tu, Tian Li et al.2021 · 26 citationsarXiv
Two Sides of Meta-Learning Evaluation: In vs. Out of Distribution2102.11503v3 · Amrith Setlur, Oscar Li, Virginia Smith2021 · 1 citationarXiv
Is Support Set Diversity Necessary for Meta-Learning?2011.14048v2 · Amrith Setlur, Oscar Li, Virginia Smith2020 · 7 citationsNeurIPS 2020 Workshop on Meta-learning
Heterogeneity for the Win: One-Shot Federated Clustering2103.00697v2 · Don Kurian Dennis, Tian Li, Virginia Smith2021 · 21 citationsarXiv
Progressive Compressed Records: Taking a Byte out of Deep Learning Data1911.00472v4 · Michael Kuchnik, George Amvrosiadis, Virginia Smith2019 · 0 citationsarXiv
A Field Guide to Federated Optimization2107.06917v1 · Jianyu Wang, Zachary Charles, Zheng Xu et al.2021 · 167 citationsarXiv
Ditto: Fair and Robust Federated Learning Through Personalization2012.04221v3 · Tian Li, Shengyuan Hu, Ahmad Beirami et al.2020 · 239 citationsarXiv
On Large-Cohort Training for Federated Learning2106.07820v1 · Zachary Charles, Zachary Garrett, Zhouyuan Huo et al.2021 · 3 citationsarXiv
Tilted Empirical Risk Minimization2007.01162v2 · Tian Li, Ahmad Beirami, Maziar Sanjabi et al.2020 · 44 citationsarXiv
A Kernel Theory of Modern Data Augmentation.31777848 · Dao, Tri, Gu, Albert, Ratner, Alexander J et al.2020 · 107 citationsProceedings of machine learning research. 2019;97:1528-1537
Federated Optimization in Heterogeneous Networks1812.06127v5 · Tian Li, Anit Kumar Sahu, Manzil Zaheer et al.2018 · 436 citationsarXiv
Fair Resource Allocation in Federated Learning1905.10497v2 · Tian Li, Maziar Sanjabi, Ahmad Beirami et al.2019 · 445 citationsarXiv
FedDANE: A Federated Newton-Type Method2001.01920v1 · Tian Li, Anit Kumar Sahu, Manzil Zaheer et al.2020 · 129 citationsarXiv
LEAF: A Benchmark for Federated Settings1812.01097v3 · Sebastian Caldas, Sai Meher Karthik Duddu, Peter Wu et al.2018 · 281 citationsarXiv
Privacy for Free: Communication-Efficient Learning with Differential Privacy Using Sketches1911.00972v2 · Tian Li, Zaoxing Liu, Vyas Sekar et al.2019 · 39 citationsarXiv
Career total: 88 works. 63 are in this corpus.Showing the 50 most recent.
Profile built from the corpus for this byline.
Author records are still filling in while the Hub is in alpha. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.