mlsanityData to 7 Oct 2026

Cyber for AI at SemEval-2025 Task 4: Forgotten but Not Lost: The Balancing Act of Selective Unlearning in Large Language Models

arXiv 2503.04795 ↗First posted 2 Mar 2025cs.CL cs.AIcited by 0

On the map: LLM safety and security → Machine unlearning

Abstract (arXiv)

Large Language Models (LLMs) face significant challenges in maintaining privacy, ethics, and compliance, when sensitive or obsolete data must be selectively removed. Retraining these models from scratch is computationally infeasible, necessitating efficient alternatives. As part of the SemEval 2025 Task 4, this work focuses on the application of selective unlearning in LLMs to address this challenge. In this paper, we present our experiments and findings, primarily leveraging global weight modification to achieve an equilibrium between effectiveness of unlearning, knowledge retention, and target model's post-unlearning utility. We also detail the task-specific evaluation mechanism, results, and challenges. Our algorithms have achieved an aggregate score of 0.409 and 0.389 on the test set for 7B and 1B target models, respectively, demonstrating promising results in verifiable LLM unlearning.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Machine unlearning, one of 12 directions in LLM safety and security. Builds on 20 of its 29 references that are in the map. First posted on arXiv 2 Mar 2025.

Builds on

20 of 29 references.

Closest ancestors · 14

Same direction, newest first

  1. 2025LUME: LLM Unlearning with Multitask Evaluations20 Feb
  2. 2024Unlearning as multi-task optimization: A normalized gradient difference approach with an adaptive learning rate29 Oct
  3. 2024Simplicity Prevails: Rethinking Negative Preference Optimization for LLM Unlearning9 Oct
  4. 2024Intrinsic Test of Unlearning Using Parametric Knowledge Traces17 Jun
  5. 2024REVS: Unlearning Sensitive Information in Language Models via Rank Editing in the Vocabulary Space13 Jun
  6. 2024Reversing the Forget-Retain Objectives: An Efficient LLM Unlearning Framework from Logit Difference12 Jun
  7. 2024Negative Preference Optimization: From Catastrophic Collapse to Effective Unlearning8 Apr
  8. 2024Digital Forgetting in Large Language Models: A Survey of Unlearning Methods2 Apr
  9. 2024The Frontier of Data Erasure: Machine Unlearning for Large Language Models23 Mar
  10. 2024Guardrail Baselines for Unlearning in LLMs5 Mar
  11. 2024Dissecting Language Models: Machine Unlearning via Selective Pruning2 Mar
  12. 2024Towards Safer Large Language Models through Machine Unlearning15 Feb
2 more
  1. 2023Unlearn What You Want to Forget: Efficient Unlearning for LLMs31 Oct
  2. 2023In-Context Unlearning: Language Models as Few Shot Unlearners11 Oct

Other references · 6

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2020Measuring Massive Multitask Language Understandingcited by 7,768
  3. 2023Composing Parameter-Efficient Modules with Arithmetic Operationscited by 155
  4. 2024The rising costs of training frontier AI modelscited by 91
  5. 2023Privacy Issues in Large Language Models: A Surveycited by 57
  6. 2024Understanding the Performance and Estimating the Cost of LLM Fine-Tuningcited by 48

Built on by

Not cited yet.