Tom Goldstein
- A Technical Report on “Erasing the Invisible”: The 2024 NeurIPS Competition on Stress Testing Image Watermarks
- Dense Backpropagation Improves Training for Sparse Mixture-of-Experts
- FineGRAIN: Evaluating Failure Modes of Text-to-Image Models with Vision Language Model Judges
- Gemstones: A Model Suite for Multi-Faceted Scaling Laws
- Quantifying Cross-Modality Memorization in Vision-Language Models
- Scaling up Test-Time Compute with Latent Reasoning: A Recurrent Depth Approach
- The Common Pile v0.1: An 8TB Dataset of Public Domain and Openly Licensed Text