6 papers
Reversible Lifelong Model Editing via Semantic Routing-Based LoRA
Haihua Luo, Xuming Ran, Tommi Kärkkäinen +4
The dynamic evolution of real-world necessitates model editing within Large Language Models. While existing methods explore modular isolation or parameter-efficient strategies, the…
A Simple Efficiency Incremental Learning Framework via Vision-Language Model with Nonlinear Multi-Adapters
Haihua Luo, Xuming Ran, Jiangrong Shen +5
Incremental Learning (IL) aims to learn new tasks while preserving previously acquired knowledge. Integrating the zero-shot learning capabilities of pre-trained vision-language mod…
Representation Finetuning for Continual Learning
Haihua Luo, Xuming Ran, Tommi Kärkkäinen +5
The world is inherently dynamic, and continual learning aims to enable models to adapt to ever-evolving data streams. While pre-trained models have shown powerful performance in co…
Key-Value Pair-Free Continual Learner via Task-Specific Prompt-Prototype
Haihua Luo, Xuming Ran, Zhengji Li +6
Continual learning aims to enable models to acquire new knowledge while retaining previously learned information. Prompt-based methods have shown remarkable performance in this dom…
AVM: Towards Structure-Preserving Neural Response Modeling in the Visual Cortex Across Stimuli and Individuals
Qi Xu, Shuai Gong, Xuming Ran +2
While deep learning models have shown strong performance in simulating neural responses, they often fail to clearly separate stable visual encoding from condition-specific adaptati…
Distillation-Guided Structural Transfer for Continual Learning Beyond Sparse Distributed Memory
Huiyan Xue, Xuming Ran, Yaxin Li +4
Sparse neural systems are gaining traction for efficient continual learning due to their modularity and low interference. Architectures such as Sparse Distributed Memory Multi-Laye…