AI

Andrew Ilyas

0000-0001-7187-5240

cs.LGstat.MLcs.CVcs.CRcs.CYcs.AIcs.NEcs.CLcs.ROcs.DS

On Valency

published · living versions

Black-box Adversarial Attacks with Limited Queries and Information

with Logan Engstrom, Anish Athalye, Jessy Lin

1 version

W_r6jpz8dbv1 · current

Preprints & journals

45 papers in the corpus · 2017–2026

Pick Your Poison: Learning to Select Poison Sets for Stronger LLM Backdoor Attacks

Aashiq Muhamed, Mona T. Diab, Virginia Smith et al.

0 citations

2609.15029v12026arXiv

Data Attribution via Sketched Metadifferentiation

Yuxi Chen, Hamza Golubovic, Han Tong et al.

0 citations

2609.15044v22026arXiv

Corrupt Plans, Clean Traces: Evading Chain-of-Thought Monitoring with Plan Injection

Keertana Chidambaram, Andrew Ilyas, Vasilis Syrgkanis

0 citations

2609.15989v12026arXiv

Using AI Agents to Automate Black-Box Audits of Personalization Algorithms at Scale

Alessandro Morosini, Sarah H. Cen, Andrew Ilyas et al.

0 citations

2606.30801v12026arXiv

Justified or Just Convincing? Error Verifiability as a Dimension of LLM Quality

Xiaoyuan Zhu, Kimberly Le Truong, Riccardo Fogliato et al.

0 citations

2604.04418v22026arXiv

Large-Scale, Longitudinal Study of Large Language Models During the 2024 US Election Season

Sarah H. Cen, Andrew Ilyas, Hedi Driss et al.

2 citations

2509.18446v12025arXiv

Optimizing Canaries for Privacy Auditing with Metagradient Descent

Matteo Boglioni, Terrance Liu, Andrew Ilyas et al.

0 citations

2507.15836v22025International Conference on Learning Representations (ICLR) 2026, pp. 92994-93013

Probably Approximately Correct Labels

Emmanuel J. Candes, Andrew Ilyas, Tijana Zrnic

0 citations

2506.10908v32025arXiv

DataMIL: Selecting Data for Robot Imitation Learning with Datamodels

Shivin Dass, Alaa Khaddaj, Logan Engstrom et al.

0 citations

2505.09603v22025arXiv

AI Supply Chains: An Emerging Ecosystem of AI Actors, Products, and Services

Aspen Hopkins, Sarah H. Cen, Andrew Ilyas et al.

2 citations

2504.20185v12025Proc. AAAI/ACM Conf. AI Ethics Soc. (AIES), 8(2), 1266-1277 (2025)

MAGIC: Near-Optimal Data Attribution for Deep Learning

Andrew Ilyas, Logan Engstrom

0 citations

2504.16430v12025arXiv

Optimizing ML Training with Metagradient Descent

Logan Engstrom, Andrew Ilyas, Benjamin Chen et al.

0 citations

2503.13751v12025arXiv

Attribute-to-Delete: Machine Unlearning via Datamodel Matching

Kristian Georgiev, Roy Rinberg, Sung Min Park et al.

0 citations

2410.23232v22024arXiv

Data Debiasing with Datamodels (D3M): Improving Subgroup Robustness via Data Selection

Saachi Jain, Kimia Hamidieh, Kristian Georgiev et al.

1 citation

2406.16846v12024arXiv

Measuring Strategization in Recommendation: Users Adapt Their Behavior to Shape Future Content

Sarah H. Cen, Andrew Ilyas, Jennifer Allen et al.

7 citations

2405.05596v12024arXiv

Decomposing and Editing Predictions by Modeling Model Computation

Harshay Shah, Andrew Ilyas, Aleksander Madry

2 citations

2404.11534v12024arXiv

User Strategization and Trustworthy Algorithms

Sarah H. Cen, Andrew Ilyas, Aleksander Madry

1 citation

2312.17666v12023arXiv

Rethinking Backdoor Attacks

Alaa Khaddaj, Guillaume Leclerc, Aleksandar Makelov et al.

7 citations

2307.10163v12023arXiv

FFCV: Accelerating Training by Removing Data Bottlenecks

Guillaume Leclerc, Andrew Ilyas, Logan Engstrom et al.

36 citations

2306.12517v12023arXiv

TRAK: Attributing Model Behavior at Scale

Sung Min Park, Kristian Georgiev, Andrew Ilyas et al.

15 citations

2303.14186v22023arXiv

Raising the Cost of Malicious AI-Powered Image Editing

Hadi Salman, Alaa Khaddaj, Guillaume Leclerc et al.

26 citations

2302.06588v12023arXiv

ModelDiff: A Framework for Comparing Learning Algorithms

Harshay Shah, Sung Min Park, Andrew Ilyas et al.

9 citations

2211.12491v12022arXiv

When does Bias Transfer in Transfer Learning?

Hadi Salman, Saachi Jain, Andrew Ilyas et al.

15 citations

2207.02842v12022arXiv

What Makes A Good Fisherman? Linear Regression under Self-Selection Bias

Yeshwanth Cherapanamjeri, Constantinos Daskalakis, Andrew Ilyas et al.

4 citations

2205.03246v22022arXiv

Estimation of Standard Auction Models

Yeshwanth Cherapanamjeri, Constantinos Daskalakis, Andrew Ilyas et al.

5 citations

2205.02060v12022arXiv

Datamodels: Predicting Predictions from Training Data

Andrew Ilyas, Sung Min Park, Logan Engstrom et al.

20 citations

2202.00622v12022arXiv

3DB: A Framework for Debugging Computer Vision Models

Guillaume Leclerc, Hadi Salman, Andrew Ilyas et al.

16 citations

2106.03805v12021arXiv

Unadversarial Examples: Designing Objects for Robust Vision

Hadi Salman, Andrew Ilyas, Logan Engstrom et al.

27 citations

2012.12235v12020arXiv

Do Adversarially Robust ImageNet Models Transfer Better?

Hadi Salman, Andrew Ilyas, Logan Engstrom et al.

147 citations

2007.08489v22020arXiv

Noise or Signal: The Role of Image Backgrounds in Object Recognition

Kai Xiao, Logan Engstrom, Andrew Ilyas et al.

161 citations

2006.09994v12020arXiv

Implementation Matters in Deep Policy Gradients: A Case Study on PPO and TRPO

Logan Engstrom, Andrew Ilyas, Shibani Santurkar et al.

146 citations

2005.12729v12020arXiv

From ImageNet to Image Classification: Contextualizing Progress on Benchmarks

Dimitris Tsipras, Shibani Santurkar, Logan Engstrom et al.

75 citations

2005.11295v12020arXiv

Identifying Statistical Bias in Dataset Replication

Logan Engstrom, Andrew Ilyas, Shibani Santurkar et al.

14 citations

2005.09619v22020arXiv

Image Synthesis with a Single (Robust) Classifier

Shibani Santurkar, Dimitris Tsipras, Brandon Tran et al.

92 citations

1906.09453v22019arXiv

Adversarial Robustness as a Prior for Learned Representations

Logan Engstrom, Andrew Ilyas, Shibani Santurkar et al.

107 citations

1906.00945v22019arXiv

Adversarial Examples Are Not Bugs, They Are Features

Andrew Ilyas, Shibani Santurkar, Dimitris Tsipras et al.

940 citations

1905.02175v42019arXiv

A Closer Look at Deep Policy Gradients

Andrew Ilyas, Logan Engstrom, Shibani Santurkar et al.

38 citations

1811.02553v42018arXiv

Evaluating and Understanding the Robustness of Adversarial Logit Pairing

Logan Engstrom, Andrew Ilyas, Anish Athalye

83 citations

1807.10272v22018arXiv

Prior Convictions: Black-Box Adversarial Attacks with Bandits and Priors

Andrew Ilyas, Logan Engstrom, Aleksander Madry

261 citations

1807.07978v32018arXiv

How Does Batch Normalization Help Optimization?

Shibani Santurkar, Dimitris Tsipras, Andrew Ilyas et al.

928 citations

1805.11604v52018arXiv

Black-box Adversarial Attacks with Limited Queries and Information

Andrew Ilyas, Logan Engstrom, Anish Athalye et al.

806 citations

1804.08598v32018arXiv

The Robust Manifold Defense: Adversarial Training using Generative Models

Ajil Jalal, Andrew Ilyas, Constantinos Daskalakis et al.

111 citations

1712.09196v52017arXiv

Query-Efficient Black-box Adversarial Examples (superceded)

Andrew Ilyas, Logan Engstrom, Anish Athalye et al.

41 citations

1712.07113v22017arXiv

Extracting Syntactic Patterns from Databases

Andrew Ilyas, Joana M. F. da Trindade, Raul Castro Fernandez et al.

2 citations

1710.11528v22017arXiv

Synthesizing Robust Adversarial Examples

Anish Athalye, Logan Engstrom, Andrew Ilyas et al.

973 citations

1707.07397v32017arXiv

Career total: 66 works. 45 are in this corpus.

Profile built from the corpus for this byline.

Author records are still filling in while the Hub is in beta. If this is your page, you'll be able to claim it soon. Spot a mistake? Tell us.