Lacuna Inc. at SemEval-2025 Task 4: LoRA-Enhanced Influence-Based Unlearning for LLMs

Aleksey Kudelya, Alexander Shirnin
6/4/2025
cs.CLcs.AI

No confident code implementation found for this paper.

Abstract

This paper describes LIBU (LoRA enhanced influence-based unlearning), an algorithm to solve the task of unlearning - removing specific knowledge from a large language model without retraining from scratch and compromising its overall utility (SemEval-2025 Task 4: Unlearning sensitive content from Large Language Models). The algorithm combines classical \textit{influence functions} to remove the influence of the data from the model and \textit{second-order optimization} to stabilize the overall utility. Our experiments show that this lightweight approach is well applicable for unlearning LLMs in different kinds of task.

View on arXivView PDF

Code Implementations

No confident code match yet

We couldn't find an author-owned or strongly-evidenced community implementation for this paper. Any repos shown below are weak matches — verify before relying on them.

No code implementations found yet.

Know of an implementation? Let us know in the comments below!

Similar Papers

Unified Parameter-Efficient Unlearning for LLMs

89% match1 repo152024

UNLEARN Efficient Removal of Knowledge in Large Language Models

88% match2024

LUNE: Efficient LLM Unlearning via LoRA Fine-Tuning with Negative Examples

87% match2025

LLM Unlearning using Gradient Ratio-Based Influence Estimation and Noise Injection

86% match2025

SOUL: Unlocking the Power of Second-Order Optimization for LLM Unlearning

86% match1 repo302024

Towards Robust and Parameter-Efficient Knowledge Unlearning for LLMs

86% match1 repo142026

Cite this paper

@article{kudelya2025lacuna,
  title  = {Lacuna Inc. at SemEval-2025 Task 4: LoRA-Enhanced Influence-Based Unlearning for LLMs},
  author = {Aleksey Kudelya and Alexander Shirnin},
  year   = {2025},
  eprint = {2506.04044},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2506.04044v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2506.04044.svg)](https://codeofpaper.com/paper/2506.04044)

Discussion