# SAFE: Machine Unlearning With Shard Graphs

arXiv: https://arxiv.org/abs/2304.13169 · HTML: https://mlsanity.com/paper/2304.13169/

## Key facts

- Cited by 28 papers in the map (data through 14 Oct 2026).
- Placed in Machine unlearning, one of 12 directions in LLM safety and security.
- Builds on 39 of its 56 references that are in the map.
- First posted on arXiv 25 Apr 2023.

## Placed in

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

## Builds on (most cited first, 30 of 39)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Switch Transformers: Scaling to Trillion Parameter Models with Simple and Efficient Sparsity](https://mlsanity.com/paper/2101.03961/) (2021)
- [Prototypical Networks for Few-shot Learning](https://mlsanity.com/paper/1703.05175/) (2017)
- [Deep Learning with Differential Privacy](https://mlsanity.com/paper/1607.00133/) (2016)
- [Describing Textures in the Wild](https://mlsanity.com/paper/1311.3618/) (2013)
- [Moment Matching for Multi-Source Domain Adaptation](https://mlsanity.com/paper/1812.01754/) (2018)
- [Machine Unlearning](https://mlsanity.com/paper/1912.03817/) (2019)
- [Renyi Differential Privacy](https://mlsanity.com/paper/1702.07476/) (2017)
- [Eternal Sunshine of the Spotless Net: Selective Forgetting in Deep Networks](https://mlsanity.com/paper/1911.04933/) (2019)
- [Certified Data Removal from Machine Learning Models](https://mlsanity.com/paper/1911.03030/) (2019)
- [Making AI Forget You: Data Deletion in Machine Learning](https://mlsanity.com/paper/1907.05012/) (2019)
- [Remember What You Want to Forget: Algorithms for Machine Unlearning](https://mlsanity.com/paper/2103.03279/) (2021)
- [Semi-supervised Domain Adaptation via Minimax Entropy](https://mlsanity.com/paper/1904.06487/) (2019)
- [Descent-to-Delete: Gradient-Based Methods for Machine Unlearning](https://mlsanity.com/paper/2007.02923/) (2020)
- [Unrolling SGD: Understanding Factors Influencing Machine Unlearning](https://mlsanity.com/paper/2109.13398/) (2021)
- [Forgetting Outside the Box: Scrubbing Deep Networks of Information Accessible from Input-Output Observations](https://mlsanity.com/paper/2003.02960/) (2020)
- [Numerical Composition of Differential Privacy](https://mlsanity.com/paper/2106.02848/) (2021)
- [Adaptive Machine Unlearning](https://mlsanity.com/paper/2106.04378/) (2021)
- [Mixed-Privacy Forgetting in Deep Networks](https://mlsanity.com/paper/2012.13431/) (2020)
- [DeltaGrad: Rapid retraining of machine learning models](https://mlsanity.com/paper/2006.14755/) (2020)
- [Graph Unlearning](https://mlsanity.com/paper/2103.14991/) (2021)
- [Machine Unlearning via Algorithmic Stability](https://mlsanity.com/paper/2102.13179/) (2021)
- [The Right to be Forgotten in Federated Learning: An Efficient Realization with Rapid Retraining](https://mlsanity.com/paper/2203.07320/) (2022)
- [Federated Unlearning via Class-Discriminative Pruning](https://mlsanity.com/paper/2110.11794/) (2021)
- [Machine Unlearning: Linear Filtration for Logit-based Classifiers](https://mlsanity.com/paper/2002.02730/) (2020)
- [Pathways: Asynchronous Distributed Dataflow for ML](https://mlsanity.com/paper/2203.12533/) (2022)
- [Forget Unlearning: Towards True Data-Deletion in Machine Learning](https://mlsanity.com/paper/2210.08911/) (2022)
- [FedRecover: Recovering from Poisoning Attacks in Federated Learning using Historical Information](https://mlsanity.com/paper/2210.10936/) (2022)
- [Mixed Differential Privacy in Computer Vision](https://mlsanity.com/paper/2203.11481/) (2022)

## Built on by (most cited first, 28 of 28)

- [Exploring the Landscape of Machine Unlearning: A Comprehensive Survey and Taxonomy](https://mlsanity.com/paper/2305.06360/) (2023)
- [On Effects of Steering Latent Representation for Large Language Model Unlearning](https://mlsanity.com/paper/2408.06223/) (2024)
- [Towards Scalable Exact Machine Unlearning Using Parameter-Efficient Fine-Tuning](https://mlsanity.com/paper/2406.16257/) (2024)
- [CPR: Retrieval Augmented Generation for Copyright Protection](https://mlsanity.com/paper/2403.18920/) (2024)
- [MultiDelete for Multimodal Machine Unlearning](https://mlsanity.com/paper/2311.12047/) (2023)
- [Learning to Unlearn for Robust Machine Unlearning](https://mlsanity.com/paper/2407.10494/) (2024)
- [Training Data Protection with Compositional Diffusion Models](https://mlsanity.com/paper/2308.01937/) (2023)
- [MU-Bench: A Multitask Multimodal Benchmark for Machine Unlearning](https://mlsanity.com/paper/2406.14796/) (2024)
- [Tangent Transformers for Composition, Privacy and Removal](https://mlsanity.com/paper/2307.08122/) (2023)
- [Distill to Delete: Unlearning in Graph Networks with Knowledge Distillation](https://mlsanity.com/paper/2309.16173/) (2023)
- [A Survey of Graph Unlearning](https://mlsanity.com/paper/2310.02164/) (2023)
- [Ticketed Learning-Unlearning Schemes](https://mlsanity.com/paper/2306.15744/) (2023)
- [A Certified Unlearning Approach without Access to Source Data](https://mlsanity.com/paper/2506.06486/) (2025)
- [Towards Reliable Forgetting: A Survey on Machine Unlearning Verification](https://mlsanity.com/paper/2506.15115/) (2025)
- [Mitigating Social Biases in Language Models through Unlearning](https://mlsanity.com/paper/2406.13551/) (2024)
- [Copyright-Protected Language Generation via Adaptive Model Fusion](https://mlsanity.com/paper/2412.06619/) (2024)
- [Speech Unlearning](https://mlsanity.com/paper/2506.00848/) (2025)
- [Towards Source-Free Machine Unlearning](https://mlsanity.com/paper/2508.15127/) (2025)
- [SUV: Scalable Large Language Model Copyright Compliance with Regularized Selective Unlearning](https://mlsanity.com/paper/2503.22948/) (2025)
- [Strong Copyright Protection for Language Models via Adaptive Model Fusion](https://mlsanity.com/paper/2407.20105/) (2024)
- [Sample Selection via Contrastive Fragmentation for Noisy Label Regression](https://mlsanity.com/paper/2502.17771/) (2025)
- [Exact Federated Continual Unlearning for Ridge Heads on Frozen Foundation Models](https://mlsanity.com/paper/2603.12977/) (2026)
- [Enabling Group Fairness in Graph Unlearning via Bi-level Debiasing](https://mlsanity.com/paper/2505.09702/) (2025)
- [Inference-time Unlearning Using Conformal Prediction](https://mlsanity.com/paper/2602.03787/) (2026)
- [GraphMU: Repairing Robustness of Graph Neural Networks via Machine Unlearning](https://mlsanity.com/paper/2406.13499/) (2024)
- [Privacy-Aware Lifelong Learning](https://mlsanity.com/paper/2505.10941/) (2025)
- [SoK: Unlearnability and Unlearning for Model Dememorization](https://mlsanity.com/paper/2605.11592/) (2026)
- [Interference-Aware Multi-Task Unlearning](https://mlsanity.com/paper/2605.19042/) (2026)

## Abstract (arXiv)

We present Synergy Aware Forgetting Ensemble (SAFE), a method to adapt large models on a diverse collection of data while minimizing the expected cost to remove the influence of training samples from the trained model. This process, also known as selective forgetting or unlearning, is often conducted by partitioning a dataset into shards, training fully independent models on each, then ensembling the resulting models. Increasing the number of shards reduces the expected cost to forget but at the same time it increases inference cost and reduces the final accuracy of the model since synergistic information between samples is lost during the independent model training. Rather than treating each shard as independent, SAFE introduces the notion of a shard graph, which allows incorporating limited information from other shards during training, trading off a modest increase in expected forgetting cost with a significant increase in accuracy, all while still attaining complete removal of residual influence after forgetting. SAFE uses a lightweight system of adapters which can be trained while reusing most of the computations. This allows SAFE to be trained on shards an order-of-magnitude smaller than current state-of-the-art methods (thus reducing the forgetting costs) while also maintaining high accuracy, as we demonstrate empirically on fine-grained computer vision datasets.