# Challenging Forgets: Unveiling the Worst-Case Forget Sets in Machine Unlearning

arXiv: https://arxiv.org/abs/2403.07362 · HTML: https://mlsanity.com/paper/2403.07362/

## Key facts

- Cited by 44 papers in the map (data through 14 Oct 2026).
- Placed in Machine unlearning, one of 12 directions in LLM safety and security.
- Builds on 63 of its 82 references that are in the map.
- First posted on arXiv 12 Mar 2024.

## Placed in

- LLM safety and security / [Machine unlearning](https://mlsanity.com/safety/machine-unlearning/)

## Builds on (most cited first, 30 of 63)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Very Deep Convolutional Networks for Large-Scale Image Recognition](https://mlsanity.com/paper/1409.1556/) (2014)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [Convex optimization](https://mlsanity.com/paper/2106.01946/) (2021)
- [Membership Inference Attacks against Machine Learning Models](https://mlsanity.com/paper/1610.05820/) (2016)
- [Machine Unlearning](https://mlsanity.com/paper/1912.03817/) (2019)
- [Membership Inference Attacks From First Principles](https://mlsanity.com/paper/2112.03570/) (2021)
- [Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networks](https://mlsanity.com/paper/1911.04933/) (2019)
- [Estimating Training Data Influence by Tracing Gradient Descent](https://mlsanity.com/paper/2002.08484/) (2020)
- [signSGD: Compressed Optimisation for Non-Convex Problems](https://mlsanity.com/paper/1802.04434/) (2018)
- [Erasing Concepts from Diffusion Models](https://mlsanity.com/paper/2303.07345/) (2023)
- [Certified Data Removal from Machine Learning Models](https://mlsanity.com/paper/1911.03030/) (2019)
- [Deep Learning on a Data Diet: Finding Important Examples Early in Training](https://mlsanity.com/paper/2107.07075/) (2021)
- [Making AI Forget You: Data Deletion in Machine Learning](https://mlsanity.com/paper/1907.05012/) (2019)
- [Unified Concept Editing in Diffusion Models](https://mlsanity.com/paper/2308.14761/) (2023)
- [SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation](https://mlsanity.com/paper/2310.12508/) (2023)
- [Who's Harry Potter? Approximate Unlearning in LLMs](https://mlsanity.com/paper/2310.02238/) (2023)
- [Remember What You Want to Forget: Algorithms for Machine Unlearning](https://mlsanity.com/paper/2103.03279/) (2021)
- [Selection via Proxy: Efficient Data Selection for Deep Learning](https://mlsanity.com/paper/1906.11829/) (2019)
- [Amnesiac Machine Learning](https://mlsanity.com/paper/2010.10981/) (2020)
- [Ablating Concepts in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2303.13516/) (2023)
- [A Survey of Machine Unlearning](https://mlsanity.com/paper/2209.02299/) (2022)
- [Descent-to-Delete: Gradient-Based Methods for Machine Unlearning](https://mlsanity.com/paper/2007.02923/) (2020)
- [An Investigation of Why Overparameterization Exacerbates Spurious Correlations](https://mlsanity.com/paper/2005.04345/) (2020)
- [Large Language Model Unlearning](https://mlsanity.com/paper/2310.10683/) (2023)
- [Forget-Me-Not: Learning to Forget in Text-to-Image Diffusion Models](https://mlsanity.com/paper/2303.17591/) (2023)
- [Systematic Evaluation of Privacy Risks of Machine Learning Models](https://mlsanity.com/paper/2003.10595/) (2020)
- [Unrolling SGD: Understanding Factors Influencing Machine Unlearning](https://mlsanity.com/paper/2109.13398/) (2021)
- [Towards Unbounded Machine Unlearning](https://mlsanity.com/paper/2302.09880/) (2023)

## Built on by (most cited first, 30 of 44)

- [Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning](https://mlsanity.com/paper/2410.07163/) (2024)
- [Defensive Unlearning with Adversarial Training for Robust Concept Erasure in Diffusion Models](https://mlsanity.com/paper/2405.15234/) (2024)
- [SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning](https://mlsanity.com/paper/2404.18239/) (2024)
- [What makes unlearning hard and what to do about it](https://mlsanity.com/paper/2406.01257/) (2024)
- [Catastrophic Failure of LLM Unlearning via Quantization](https://mlsanity.com/paper/2410.16454/) (2024)
- [Towards LLM Unlearning Resilient to Relearning Attacks: A Sharpness-Aware Minimization Perspective and Beyond](https://mlsanity.com/paper/2502.05374/) (2025)
- [LLM Unlearning via Loss Adjustment with Only Forget Data](https://mlsanity.com/paper/2410.11143/) (2024)
- [Are we making progress in unlearning? Findings from the first NeurIPS unlearning competition](https://mlsanity.com/paper/2406.09073/) (2024)
- [Exploring Criteria of Loss Reweighting to Enhance LLM Unlearning](https://mlsanity.com/paper/2505.11953/) (2025)
- [WAGLE: Strategic Weight Attribution for Effective and Modular Unlearning in Large Language Models](https://mlsanity.com/paper/2410.17509/) (2024)
- [MU-Bench: A Multitask Multimodal Benchmark for Machine Unlearning](https://mlsanity.com/paper/2406.14796/) (2024)
- [MUNBa: Machine Unlearning via Nash Bargaining](https://mlsanity.com/paper/2411.15537/) (2024)
- [Tool Unlearning for Tool-Augmented LLMs](https://mlsanity.com/paper/2502.01083/) (2025)
- [GUARD: Generation-time LLM Unlearning via Adaptive Restriction and Detection](https://mlsanity.com/paper/2505.13312/) (2025)
- [Decoupling the Class Label and the Target Concept in Machine Unlearning](https://mlsanity.com/paper/2406.08288/) (2024)
- [LLM Unlearning Reveals a Stronger-Than-Expected Coreset Effect in Current Benchmarks](https://mlsanity.com/paper/2504.10185/) (2025)
- [Towards Natural Machine Unlearning](https://mlsanity.com/paper/2405.15495/) (2024)
- [Towards Lifecycle Unlearning Commitment Management: Measuring Sample-level Unlearning Completeness](https://mlsanity.com/paper/2506.06112/) (2025)
- [Machine Unlearning under Retain-Forget Entanglement](https://mlsanity.com/paper/2603.26569/) (2026)
- [UIPE: Enhancing LLM Unlearning by Removing Knowledge Related to Forgetting Targets](https://mlsanity.com/paper/2503.04693/) (2025)
- [Rectifying Privacy and Efficacy Measurements in Machine Unlearning: A New Inference Attack Perspective](https://mlsanity.com/paper/2506.13009/) (2025)
- [OFFSIDE: Benchmarking Unlearning Misinformation in Multimodal Large Language Models](https://mlsanity.com/paper/2510.22535/) (2025)
- [Go Beyond Your Means: Unlearning with Per-Sample Gradient Orthogonalization](https://mlsanity.com/paper/2503.02312/) (2025)
- [CURE4Rec: A Benchmark for Recommendation Unlearning with Deeper Influence](https://mlsanity.com/paper/2408.14393/) (2024)
- [MUBox: A Critical Evaluation Framework of Deep Machine Unlearning](https://mlsanity.com/paper/2505.08576/) (2025)
- [Not All Tokens Are Meant to Be Forgotten](https://mlsanity.com/paper/2506.03142/) (2025)
- [SUV: Scalable Large Language Model Copyright Compliance with Regularized Selective Unlearning](https://mlsanity.com/paper/2503.22948/) (2025)
- [FROG: Fair Removal on Graphs](https://mlsanity.com/paper/2503.18197/) (2025)
- [Leveraging Per-Instance Privacy for Machine Unlearning](https://mlsanity.com/paper/2505.18786/) (2025)
- [Tackling Fake Forgetting through Uncertainty Quantification](https://mlsanity.com/paper/2501.19403/) (2025)

## Abstract (arXiv)

The trustworthy machine learning (ML) community is increasingly recognizing the crucial need for models capable of selectively 'unlearning' data points after training. This leads to the problem of machine unlearning (MU), aiming to eliminate the influence of chosen data points on model performance, while still maintaining the model's utility post-unlearning. Despite various MU methods for data influence erasure, evaluations have largely focused on random data forgetting, ignoring the vital inquiry into which subset should be chosen to truly gauge the authenticity of unlearning performance. To tackle this issue, we introduce a new evaluative angle for MU from an adversarial viewpoint. We propose identifying the data subset that presents the most significant challenge for influence erasure, i.e., pinpointing the worst-case forget set. Utilizing a bi-level optimization principle, we amplify unlearning challenges at the upper optimization level to emulate worst-case scenarios, while simultaneously engaging in standard training and unlearning at the lower level, achieving a balance between data influence erasure and model utility. Our proposal offers a worst-case evaluation of MU's resilience and effectiveness. Through extensive experiments across different datasets (including CIFAR-10, 100, CelebA, Tiny ImageNet, and ImageNet) and models (including both image classifiers and generative models), we expose critical pros and cons in existing (approximate) unlearning strategies. Our results illuminate the complex challenges of MU in practice, guiding the future development of more accurate and robust unlearning algorithms. The code is available at https://github.com/OPTML-Group/Unlearn-WorstCase.