Tom Goldstein's Lab at University of Maryland, College Park

university

http://www.cs.umd.edu/~tomg/

AI & ML interests

AI security & privacy, algorithmic bias, foundations of ML

Recent Activity

JonasGeiping authored a paper 6 days ago

Answer Matching Outperforms Multiple Choice for Language Model Evaluation

JonasGeiping authored a paper 6 days ago

MedSAMix: A Training-Free Model Merging Approach for Medical Image Segmentation

JonasGeiping authored a paper 6 days ago

The Illusion of Diminishing Returns: Measuring Long Horizon Execution in LLMs

View all activity

Papers

Teaching Pretrained Language Models to Think Deeper with Retrofitted Recurrence

Gemstones: A Model Suite for Multi-Faceted Scaling Laws

View all Papers