OpenUnlearning: Accelerating LLM Unlearning via Unified Benchmarking of Methods and Metrics

TL;DR

This paper introduces OpenUnlearning, a comprehensive framework for benchmarking LLM unlearning methods and metrics, aiming to standardize evaluation, improve reproducibility, and accelerate progress in safe model deployment.

Contribution

It presents a unified, extensible benchmarking platform that integrates multiple algorithms, evaluation metrics, and datasets for systematic analysis of LLM unlearning techniques.

Findings

OpenUnlearning includes 13 algorithms and 16 evaluation metrics.

Benchmarking reveals significant variability in unlearning effectiveness.

The framework facilitates analysis of forgetting behaviors across 450+ checkpoints.

Abstract

Robust unlearning is crucial for safely deploying large language models (LLMs) in environments where data privacy, model safety, and regulatory compliance must be ensured. Yet the task is inherently challenging, partly due to difficulties in reliably measuring whether unlearning has truly occurred. Moreover, fragmentation in current methodologies and inconsistent evaluation metrics hinder comparative analysis and reproducibility. To unify and accelerate research efforts, we introduce OpenUnlearning, a standardized and extensible framework designed explicitly for benchmarking both LLM unlearning methods and metrics. OpenUnlearning integrates 13 unlearning algorithms and 16 diverse evaluations across 3 leading benchmarks (TOFU, MUSE, and WMDP) and also enables analyses of forgetting behaviors across 450+ checkpoints we publicly release. Leveraging OpenUnlearning, we propose a novel meta-evaluation benchmark focused specifically on assessing the faithfulness and robustness of evaluation metrics themselves. We also benchmark diverse unlearning methods and provide a comparative analysis against an extensive evaluation suite. Overall, we establish a clear, community-driven pathway toward rigorous development in LLM unlearning research.