Virginia Smith

e3: Learning to Explore Enables Extrapolation of Test-Time Compute for LLMs

Learning to Reason on Hard Problems with Privileged On-Policy Exploration

Membership Inference Attacks for Unseen Classes

BLUR: A Benchmark for LLM Unlearning Robust to Forget-Retain Overlap

Unlearning or Obfuscating? Jogging the Memory of Unlearned LLMs via Benign Relearning

SAEs Can Improve Unlearning: Dynamic Sparse Autoencoder Guardrails for Precision Unlearning in LLMs

Many-Objective Multi-Solution Transport