# Class Machine Unlearning for Complex Data via Concepts Inference and Data Poisoning

arXiv: https://arxiv.org/abs/2405.15662 · HTML: https://mlsanity.com/paper/2405.15662/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Placed in Machine unlearning, one of 12 directions in LLM safety and security.
- Builds on 32 of its 50 references that are in the map.
- First posted on arXiv 24 May 2024.

## Placed in

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

## Builds on (most cited first, 30 of 32)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Axiomatic Attribution for Deep Networks](https://mlsanity.com/paper/1703.01365/) (2017)
- [Membership Inference Attacks against Machine Learning Models](https://mlsanity.com/paper/1610.05820/) (2016)
- [A Survey on Evaluation of Large Language Models](https://mlsanity.com/paper/2307.03109/) (2023)
- [Machine Unlearning](https://mlsanity.com/paper/1912.03817/) (2019)
- [Concept Bottleneck Models](https://mlsanity.com/paper/2007.04612/) (2020)
- [TOFU: A Task of Fictitious Unlearning for LLMs](https://mlsanity.com/paper/2401.06121/) (2024)
- [The WMDP Benchmark: Measuring and Reducing Malicious Use With Unlearning](https://mlsanity.com/paper/2403.03218/) (2024)
- [Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning](https://mlsanity.com/paper/2404.05868/) (2024)
- [Knowledge Unlearning for Mitigating Privacy Risks in Language Models](https://mlsanity.com/paper/2210.01504/) (2022)
- [Efficient Processing of Deep Neural Networks: A Tutorial and Survey](https://mlsanity.com/paper/1703.09039/) (2017)
- [A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects](https://mlsanity.com/paper/2004.02806/) (2020)
- [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)
- [Amnesiac Machine Learning](https://mlsanity.com/paper/2010.10981/) (2020)
- [Large Language Model Unlearning](https://mlsanity.com/paper/2310.10683/) (2023)
- [Post-hoc Concept Bottleneck Models](https://mlsanity.com/paper/2205.15480/) (2022)
- [In-Context Unlearning: Language Models as Few Shot Unlearners](https://mlsanity.com/paper/2310.07579/) (2023)
- [Graph Unlearning](https://mlsanity.com/paper/2103.14991/) (2021)
- [Boundary Unlearning](https://mlsanity.com/paper/2303.11570/) (2023)
- [Federated Unlearning via Class-Discriminative Pruning](https://mlsanity.com/paper/2110.11794/) (2021)
- [Guided Integrated Gradients: An Adaptive Path Method for Removing Noise](https://mlsanity.com/paper/2106.09788/) (2021)
- [Machine Unlearning: Linear Filtration for Logit-based Classifiers](https://mlsanity.com/paper/2002.02730/) (2020)
- [Recommendation Unlearning](https://mlsanity.com/paper/2201.06820/) (2022)
- [Learn to Forget: Machine Unlearning via Neuron Masking](https://mlsanity.com/paper/2003.10933/) (2020)
- [Machine Unlearning via Null Space Calibration](https://mlsanity.com/paper/2404.13588/) (2024)
- [Reinforcement Unlearning](https://mlsanity.com/paper/2312.15910/) (2023)
- [The Frontier of Data Erasure: Machine Unlearning for Large Language Models](https://mlsanity.com/paper/2403.15779/) (2024)

## Built on by (most cited first, 6 of 6)

- [A Systematic Review of Poisoning Attacks Against Large Language Models](https://mlsanity.com/paper/2506.06518/) (2025)
- [Zero-shot Class Unlearning via Layer-wise Relevance Analysis and Neuronal Path Perturbation](https://mlsanity.com/paper/2410.23693/) (2024)
- [Federated TrustChain: Blockchain-Enhanced LLM Training and Unlearning](https://mlsanity.com/paper/2406.04076/) (2024)
- [Exploring and Leveraging Class Vectors for Classifier Editing](https://mlsanity.com/paper/2510.11268/) (2025)
- [From Thinking to Output: Chain-of-Thought and Text Generation Characteristics in Reasoning Language Models](https://mlsanity.com/paper/2506.21609/) (2025)
- [MeGU: Machine-Guided Unlearning with Target Feature Disentanglement](https://mlsanity.com/paper/2602.17088/) (2026)

## Abstract (arXiv)

Machine unlearning aims to remove the influence of specified training data or knowledge from a trained model without requiring full retraining. This capability is particularly important for modern image classifiers and large language models (LLMs), where retraining can be computationally expensive. However, machine unlearning on complex data remains difficult because the target information is often distributed across multiple semantic elements. Existing methods mainly remove samples, modify labels, or edit model parameters to reduce the influence of the forgetting target. These approaches usually do not explicitly identify which semantic concepts connect the forgetting target to the model's prediction or generated response. As a result, it is difficult to determine which information to modify. This uncertainty may leave residual target information or unnecessarily affect knowledge that should be retained. To address this gap, we propose a concept-guided poisoning unlearning framework that explicitly identifies the concepts that contribute most strongly to the target class or knowledge and uses them to guide the model update. For image classification, our method first identifies class-relevant concepts with a Post-hoc Concept Bottleneck Model, localizes the image regions that express these concepts, and constructs replacement-based poisoned samples. For LLMs, it elicits the target knowledge through multiple questions, aggregates Integrated Gradients across the resulting responses to identify consistently important content, and masks this content to construct poisoned training targets. Experiments on multiple datasets show that the proposed framework achieves effective unlearning across different tasks while largely preserving retained model utility.