164 citations · 1.3k across the 73 of their papers we have counts for
51 papers · 1 filter
TANGO: Watermarking Masked Diffusion Language Models in Token Pairs
Kasra Arabi, Nir Weinberger, Micah Goldblum +1
Masked-diffusion language models fill in masked positions in parallel and in no fixed order. Most practical text watermarks assume left-to-right generation. They key each token to…
Cascade: Token-Sharded Private LLM Inference
Rahul Thomas, Louai Zahran, Erica Choi +3
As LLMs continue to increase in parameter size, the computational resources required to run them are available to fewer parties. Therefore, third-party inference services -- where…
Small Batch Size Training for Language Models: When Vanilla SGD Works, and Why Gradient Accumulation Is Wasteful
Martin Marek, Sanae Lotfi, Aditya Somasundaram +2
Conventional wisdom dictates that small batch sizes make language model pretraining and fine-tuning unstable, motivating gradient accumulation, which trades off the number of optim…
Identifying and Evaluating Inactive Heads in Pretrained LLMs
Pedro Sandoval-Segura, Xijun Wang, Ashwinee Panda +4
Attention is foundational to large language models (LLMs), enabling different heads to have diverse focus on relevant input tokens. However, learned behaviors like attention sinks,…
Commercial LLM Agents Are Already Vulnerable to Simple Yet Dangerous Attacks
Ang Li, Yin Zhou, Vethavikashini Chithrra Raghuram +2
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 in…
Gemstones: A Model Suite for Multi-Faceted Scaling Laws
Sean McLeish, John Kirchenbauer, David Yu Miller +5
Scaling laws are typically fit using a family of models with a narrow range of frozen hyperparameter choices. In this work we study scaling laws using multiple architectural shapes…
Refusal Tokens: A Simple Way to Calibrate Refusals in Large Language Models
Neel Jain, Aditya Shrivastava, Chenyang Zhu +6
A key component of building safe and reliable language models is enabling the models to appropriately refuse to follow certain instructions or answer certain questions. We may want…
A Simple Baseline for Predicting Events with Auto-Regressive Tabular Transformers
Alex Stein, Samuel Sharpe, Doron Bergman +5
Many real-world applications of tabular data involve using historic events to predict properties of new ones, for example whether a credit card transaction is fraudulent or what ra…
Just How Flexible are Neural Networks in Practice?
Ravid Shwartz-Ziv, Micah Goldblum, Arpit Bansal +3
It is widely believed that a neural network can fit a training set containing at least as many samples as it has parameters, underpinning notions of overparameterized and underpara…
Generating Potent Poisons and Backdoors from Scratch with Guided Diffusion
Hossein Souri, Arpit Bansal, Hamid Kazemi +7
Modern neural networks are often trained on massive datasets that are web scraped with minimal human inspection. As a result of this insecure curation pipeline, an adversary can po…
A Performance-Driven Benchmark for Feature Selection in Tabular Deep Learning
Valeriia Cherepanova, Roman Levin, Gowthami Somepalli +5
Academic tabular benchmarks often contain small sets of curated features. In contrast, data scientists typically collect as many features as possible into their datasets, and even…
Baseline Defenses for Adversarial Attacks Against Aligned Language Models
Neel Jain, Avi Schwarzschild, Yuxin Wen +7
As Large Language Models quickly become ubiquitous, it becomes critical to understand their security vulnerabilities. Recent work shows that text optimizers can produce jailbreakin…
On the Reliability of Watermarks for Large Language Models
John Kirchenbauer, Jonas Geiping, Yuxin Wen +7
As LLMs become commonplace, machine-generated text has the potential to flood the internet with spam, social media bots, and valueless content. Watermarking is a simple and effecti…
Understanding and Mitigating Copying in Diffusion Models
Gowthami Somepalli, Vasu Singla, Micah Goldblum +2
Images generated by diffusion models like Stable Diffusion are increasingly widespread. Recent works and even lawsuits have shown that these models are prone to replicating their t…
What Can We Learn from Unlearnable Datasets?
Pedro Sandoval-Segura, Vasu Singla, Jonas Geiping +2
In an era of widespread web scraping, unlearnable dataset methods have the potential to protect data privacy by preventing deep neural networks from generalizing. But in addition t…
A Cookbook of Self-Supervised Learning
Randall Balestriero, Mark Ibrahim, Vlad Sobal +16
Self-supervised learning, dubbed the dark matter of intelligence, is a promising path to advance machine learning. Yet, much like cooking, training SSL methods is a delicate art wi…
Hard Prompts Made Easy: Gradient-Based Discrete Optimization for Prompt Tuning and Discovery
Yuxin Wen, Neel Jain, John Kirchenbauer +3
The strength of modern generative models lies in their ability to be controlled through text-based prompts. Typical "hard" prompts are made from interpretable words and tokens, and…
Exploring and Exploiting Decision Boundary Dynamics for Adversarial Robustness
Yuancheng Xu, Yanchao Sun, Micah Goldblum +2
The robustness of a deep classifier can be characterized by its margins: the decision boundary's distances to natural data points. However, it is unclear whether existing robust tr…
Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Models
Gowthami Somepalli, Vasu Singla, Micah Goldblum +2
Cutting-edge diffusion models produce images with high quality and customizability, enabling them to be used for commercial art and graphic design purposes. But do diffusion models…
K-SAM: Sharpness-Aware Minimization at the Speed of SGD
Renkun Ni, Ping-yeh Chiang, Jonas Geiping +3
Sharpness-Aware Minimization (SAM) has recently emerged as a robust technique for improving the accuracy of deep neural networks. However, SAM incurs a high computational cost in p…
Thinking Two Moves Ahead: Anticipating Other Users Improves Backdoor Attacks in Federated Learning
Yuxin Wen, Jonas Geiping, Liam Fowl +4
Federated learning is particularly susceptible to model poisoning and backdoor attacks because individual users have direct control over the training data and model updates. At the…
Canary in a Coalmine: Better Membership Inference with Ensembled Adversarial Queries
Yuxin Wen, Arpit Bansal, Hamid Kazemi +4
As industrial applications are increasingly automated by machine learning models, enforcing personal data ownership and intellectual property rights requires tracing training data…
How Much Data Are Augmentations Worth? An Investigation into Scaling Laws, Invariance, and Implicit Regularization
Jonas Geiping, Micah Goldblum, Gowthami Somepalli +3
Despite the clear performance benefits of data augmentations, little is known about why they are so effective. In this paper, we disentangle several key mechanisms through which da…
Transfer Learning with Deep Tabular Models
Roman Levin, Valeriia Cherepanova, Avi Schwarzschild +5
Recent work on deep learning for tabular data demonstrates the strong performance of deep tabular models, often bridging the gap between gradient boosted decision trees and neural…