Versatile Incremental Learning: Towards Class and Domain-Agnostic Incremental Learning

MY Park, JH Lee, GM Park - European Conference on Computer Vision, 2024 - Springer
Incremental Learning (IL) aims to accumulate knowledge from sequential input tasks while
overcoming catastrophic forgetting. Existing IL methods typically assume that an incoming …

Online Continuous Generalized Category Discovery

KH Park, H Lee, K Song, GM Park - European Conference on Computer …, 2024 - Springer
With the advancement of deep neural networks in computer vision, artificial intelligence (AI)
is widely employed in real-world applications. However, AI still faces limitations in mimicking …

MUNBa: Machine Unlearning via Nash Bargaining

J Wu, M Harandi - arxiv preprint arxiv:2411.15537, 2024 - arxiv.org
Machine Unlearning (MU) aims to selectively erase harmful behaviors from models while
retaining the overall utility of the model. As a multi-task learning problem, MU involves …

[PDF][PDF] Advancing Unlearning in Generative AI: Toward Responsible Artificial General Intelligence

Y Zhao, H Du, Y Lin, D Niyato, HV Poor - researchgate.net
The rapid advances in generative artificial intelligence (GenAI) have revolutionized AI
applications, raising critical concerns regarding privacy, ethical use, and data ownership …

No Training Data, No Cry: Model Editing without Training Data or Fine-tuning

D Kashyap, T Narshana, C Murti, C Bhattacharyya - openreview.net
Model Editing (ME)--such as classwise unlearning and structured pruning--is a nascent field
that deals with identifying editable components that, when modified, significantly change the …