---
license: apache-2.0
language:
- en
tags:
- recommender-system
- machine-unlearning
- benchmark
pretty_name: ERASE
size_categories:
- 1K<n<10K
---

# ERASE – A Real-World Aligned Benchmark for Unlearning in Recommender Systems

ERASE is a benchmark for evaluating machine unlearning methods in recommender systems under real-world inspired conditions. It provides a standardized framework for measuring unlearning **efficiency**, **utility**, and **effectiveness** across multiple datasets and models.

For full details, code, and documentation, see the [GitHub repository](https://github.com/deem-data/erase-bench/).

## Repository Contents

- **`saved/`** — Trained, retrained, and unlearned model checkpoints
- **`logs/`** — Logs from training, retraining, and unlearning runs, including efficiency, utility, and effectiveness metrics
- **`dataset/`** — Instructions and scripts for downloading the datasets used in the benchmark