Lacuna Inc. at SemEval-2025 Task 4: LoRA-Enhanced Influence-Based Unlearning for LLMs
On the map: LLM safety and security → Machine unlearning
Abstract (arXiv)
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 textitinfluence functions to remove the influence of the data from the model and textitsecond-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.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Placed in Machine unlearning, one of 12 directions in LLM safety and security. Builds on 9 of its 13 references that are in the map. First posted on arXiv 4 Jun 2025.
Builds on
9 of 13 references.
Closest ancestors · 3
Same direction, newest first
- 2025LUME: LLM Unlearning with Multitask Evaluations20 Feb
- 2024MUSE: Machine Unlearning Six-Way Evaluation for Language Models8 Jul
- 2023Fast Machine Unlearning Without Retraining Through Selective Synaptic Dampening15 Aug
Other references · 6
- 2014Adam: A Method for Stochastic Optimizationcited by 36,960
- 2020Measuring Massive Multitask Language Understandingcited by 7,768
- 2017Understanding Black-box Predictions via Influence Functionscited by 1,714
- 2019Machine Unlearningcited by 1,020
- 2024OLMo: Accelerating the Science of Language Modelscited by 599
- 2023Sophia: A Scalable Stochastic Second-order Optimizer for Language Model Pre-trainingcited by 260
Built on by
Citing papers per year: 2026: 1.
1 paper.