mlsanityData to 14 Oct 2026

Toward Efficient Data-Free Unlearning

arXiv 2412.13790 ↗First posted 18 Dec 2024cs.LGcited by 4

On the map: LLM safety and security → Machine unlearning

Abstract (arXiv)

Machine unlearning without access to real data distribution is challenging. The existing method based on data-free distillation achieved unlearning by filtering out synthetic samples containing forgetting information but struggled to distill the retaining-related knowledge efficiently. In this work, we analyze that such a problem is due to over-filtering, which reduces the synthesized retaining-related information. We propose a novel method, Inhibited Synthetic PostFilter (ISPF), to tackle this challenge from two perspectives: First, the Inhibited Synthetic, by reducing the synthesized forgetting information; Second, the PostFilter, by fully utilizing the retaining-related information in synthesized samples. Experimental results demonstrate that the proposed ISPF effectively tackles the challenge and outperforms existing methods.

Key facts. Cited by 4 papers in the map (data through 14 Oct 2026). Placed in Machine unlearning, one of 12 directions in LLM safety and security. Builds on 17 of its 25 references that are in the map. First posted on arXiv 18 Dec 2024.

Builds on

17 of 25 references.

Closest ancestors · 5

Same direction, newest first

  1. 2024Machine Unlearning via Null Space Calibration21 Apr
  2. 2023SalUn: Empowering Machine Unlearning via Gradient-based Weight Saliency in Both Image Classification and Generation19 Oct
  3. 2023Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampening15 Aug
  4. 2023Boundary Unlearning21 Mar
  5. 2021Unrolling SGD: Understanding Factors Influencing Machine Unlearning27 Sep

Other references · 12

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2016Membership Inference Attacks against Machine Learning Modelscited by 2,281
  3. 2014Striving for Simplicity: The All Convolutional Netcited by 990
  4. 2019Data-Free Learning of Student Networkscited by 204
  5. 2019Zero-shot Knowledge Transfer via Adversarial Belief Matchingcited by 129
  6. 2019Zero-Shot Knowledge Distillation in Deep Networkscited by 121
  7. 2019Data-Free Adversarial Distillationcited by 90
  8. 2020Data-Free Network Quantization With Adversarial Knowledge Distillationcited by 80
  9. 2021Contrastive Model Inversion for Data-Free Knowledge Distillationcited by 66
  10. 2020Formalizing Data Deletion in the Context of the Right to be Forgottencited by 50
  11. 2022Momentum Adversarial Distillation: Handling Large Distribution Shifts in Data-Free Knowledge Distillationcited by 31
  12. 2024Small Scale Data-Free Knowledge Distillationcited by 20

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 2; 2026: 2.

4 papers.

  1. 2026Robust Continual Unlearning against Knowledge Erosion and Forgetting Reversalcited by 1
  2. 2025Task-Agnostic Federation over Decentralized Data: Research Landscape and Visionscited by 0
  3. 2025Machine Unlearning for Streaming Forgettingcited by 0
  4. 2026SPACE: Source-free Proxy Anchor Concept Erasure for MLLMscited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Remaining-data-free Machine Unlearning by Suppressing Sample Contribution
  2. 2024Towards Natural Machine Unlearning
  3. 2025Towards Source-Free Machine Unlearning
  4. 2025Decoupled Distillation to Erase: A General Unlearning Method for Any Class-centric Tasks
  5. 2025Synthetic Forgetting without Access: A Few-shot Zero-glance Framework for Machine Unlearning
  6. 2024An Information Theoretic Approach to Machine Unlearning
  7. 2024Efficient Knowledge Deletion from Trained Models through Layer-wise Partial Machine Unlearning
  8. 2026SHRED: Retain-Set-Free Unlearning via Self-Distillation with Logit Demotion