LLM Unlearning Benchmarks are Weak Measures of Progress – Machine Learning Blog | ML@CMU
TL;DR: “Machine unlearning” aims to remove data from models without retraining the model completely. Unfortunately, state-of-the-art ...
Read moreTag: MLCMU
Copilot Arena: A Platform for Code – Machine Learning Blog | ML@CMU
Figure 1. Copilot Arena is a VSCode extension that collects human preferences of code directly from ...
Read more
Optimizing LLM Test-Time Compute Involves Solving a Meta-RL Problem – Machine Learning Blog | ML@CMU
Figure 1: Training models to optimize test-time compute and learn “how to discover” correct responses, as ...
Read more