Micah Goldblum

Just How Toxic is Data Poisoning? A Unified Benchmark for Backdoor and Data Poisoning Attacks

2020 · arXiv (Cornell University)

Data poisoning and backdoor attacks manipulate training data in order to cause models to fail during inference. A recent survey of industry practitioners found that data poisoning is the number one concern among threats ranging …

Data Augmentation for Meta-Learning

2020 · arXiv (Cornell University)

Conventional image classifiers are trained by randomly sampling mini-batches of images. To achieve state-of-the-art performance, practitioners use sophisticated data augmentation schemes to expand the amount of training data available for sampling. In contrast, meta-learning algorithms …

Can Neural Nets Learn the Same Model Twice? Investigating Reproducibility and Double Descent from the Decision Boundary Perspective

2022 · arXiv (Cornell University)

We discuss methods for visualizing neural network decision boundaries and decision regions. We use these visualizations to investigate issues related to reproducibility and generalization in neural network training. We observe that changes in model architecture …

Fishing for User Data in Large-Batch Federated Learning via Gradient Magnification

2022 · arXiv (Cornell University)

Federated learning (FL) has rapidly risen in popularity due to its promise of privacy and efficiency. Previous works have exposed privacy vulnerabilities in the FL pipeline by recovering user data from gradient updates. However, existing …

Can You Learn an Algorithm? Generalizing from Easy to Hard Problems with\n Recurrent Networks

2021 · arXiv (Cornell University)

Deep neural networks are powerful machines for visual pattern recognition,\nbut reasoning tasks that are easy for humans may still be difficult for neural\nmodels. Humans possess the ability to extrapolate reasoning strategies learned\non simple problems to …

Seeing in Words: Learning to Classify through Language Bottlenecks

2023 · arXiv (Cornell University)

Neural networks for computer vision extract uninterpretable features despite achieving high accuracy on benchmarks. In contrast, humans can explain their predictions using succinct and intuitive descriptions. To incorporate explainability into neural networks, we train a …

Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks

2025 · arXiv (Cornell University)

A high volume of recent ML security literature focuses on attacks against aligned large language models (LLMs). These attacks may extract private information or coerce the model into producing harmful outputs. In real-world deployments, LLMs …