paper-with-me

Papers

SPARC: Subspace-Aware Prompt Adaptation for Robust Continual Learning in LLMs

2025-02-05 · Dinithi Jayasuriya, Sina Tayebati, Davide Ettori, Ranganath Krishnan, Amit Ranjan Trivedi

We propose SPARC, a lightweight continual learning framework for large language models (LLMs) that enables efficient task adaptation through prompt tuning in a lower-dimensional space. By leveraging principal component analysis (PCA), we identify a compact subspace of the training data. Optimizing prompts in this lower-dimensional space enhances training efficiency, as it focuses updates on the most relevant features while reducing computational overhead. Furthermore, since the model's internal structure remains unaltered, the extensive knowledge gained from pretraining is fully preserved, ensuring that previously learned information is not compromised during adaptation. Our method achieves high knowledge retention in both task-incremental and domain-incremental continual learning setups while fine-tuning only 0.04% of the model's parameters. Additionally, by integrating LoRA, we enhance adaptability to computational constraints, allowing for a tradeoff between accuracy and training cost. Experiments on the SuperGLUE benchmark demonstrate that our PCA-based prompt tuning combined with LoRA maintains full knowledge retention while improving accuracy, utilizing only 1% of the model's parameters. These results establish our approach as a scalable and resource-efficient solution for continual learning in LLMs.

📄 PDF Abstract BibTeX arXiv:2502.02909

Code (0)

등록된 구현이 없습니다.

Tasks

Continual Learning

Similar Papers 제목 키워드 기반

SparCL: Sparse Continual Learning on the Edge

2022-09-20 · Zifeng Wang, Zheng Zhan, Yifan Gong, Geng Yuan 외

Existing work in continual learning (CL) focuses on mitigating catastrophic forgetting, i.e., model performance deterioration on past tasks when learning a new task. However, the training efficiency of a CL system is und…

Continual Learning

SPARCL: Spectral Partitioned Analytic Continual Learning

2026-08-21 · James Hartley, Zeropy Surio, Daniel Whitmore, Hannah Clarke 외 arxiv

Analytic continual learning has emerged as a strong exemplar-free alternative to gradient-based class-incremental learning because it replaces iterative optimization with closed-form ridge updates. Yet the usual forgetti…

class-incremental learningContinual Learning

Subspace Optimization for Backpropagation-Free Continual Test-Time Adaptation

2026-03-30 · Damian Sójka, Sebastian Cygert, Marc Masana arxiv

We introduce PACE, a backpropagation-free continual test-time adaptation system that directly optimizes the affine parameters of normalization layers. Existing derivative-free approaches struggle to balance runtime effic…

Test-time Adaptation

Geo-LoRA: Geometry-Aware Subspace Evolution for Low-Rank Adaptation in Continual Learning

2026-08-27 · Yibo Feng arxiv

Rehearsal-free class-incremental learning (CIL) with LoRA adapters remains challenging because the low-rank subspaces updated across tasks evolve without geometric control, causing unstable shared representations and rep…

class-incremental learningContinual Learning

Unifying Detection and Adaptation in Task-Free Continual Learning

2026-08-27 · Dezheng Han, Anbang Zhang, Zhihao Zhu, Shuaishuai Guo arxiv

To mitigate catastrophic forgetting in downstream continual learning (CL) for large language models (LLMs), existing methods typically constrain parameter updates or introduce task-specific adaptation modules. However, t…

Continual Learning