Parameter Importance-Driven Continual Learning for Foundation Models
cs.LG, cs.AI
Submitted: 2025-11-19
Updated: 2026-09-23
Code: https://github.com/wanglingxiang0717/PIECE
Terminology
Sources
- Recurrent Knowledge Identification and Fusion for Language Model Continual Learning
- Parameter Efficient Continual Learning with Dynamic Low-Rank Adaptation
- Evaluating Large Language Models Trained on Code
- Revisiting Natural Gradient for Deep Networks
- DATA: Decomposed Attention-based Task Adaptation for Rehearsal-Free Continual Learning
- Theoretical Insights into Fine-Tuning Attention Mechanism: Generalization and Optimization
- Parameter-Efficient Fine-Tuning for Foundation Models
- Continual Learning with Strong Experience Replay
- LLaMA: Open and Efficient Foundation Language Models
- TRACE: A Comprehensive Benchmark for Continual Learning in Large Language Models
- Position: Multimodal Large Language Models Can Significantly Advance Scientific Reasoning
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks