parameter importance

Parameter Importance is Not Static: Evolving Parameter Isolation for Supervised Fine-Tuning ACL 2026 Adapting In-Domain Few-Shot Segmentation to New Domains without Source Domain Retraining ICCV 2025 Dynamic Fisher-weighted Model Merging via Bayesian Optimization NAACL 2025 Learn from Downstream and Be Yourself in Multimodal Large Language Models Fine-Tuning ICML 2025 Math Neurosurgery: Isolating Language Models’ Math Reasoning Abilities Using Only Forward Passes ACL 2025 Optimal Brain Apoptosis ICLR 2025 Recurrent Knowledge Identification and Fusion for Language Model Continual Learning ACL 2025 Cherry on Top: Parameter Heterogeneity and Quantization in Large Language Models NeurIPS 2024 FIARSE: Model-Heterogeneous Federated Learning via Importance-Aware Submodel Extraction NeurIPS 2024 Parameter Disparities Dissection for Backdoor Defense in Heterogeneous Federated Learning NeurIPS 2024 Surrogate Active Subspaces for Jump-Discontinuous Functions AISTATS 2024 Pruning Pre-trained Language Models with Principled Importance and Self-regularization ACL 2023 Numerical Optimizations for Weighted Low-rank Estimation on Language Models EMNLP 2022 Unifying Importance Based Regularisation Methods for Continual Learning AISTATS 2022 You Only Derive Once (YODO): Automatic Differentiation for Efficient Sensitivity Analysis in Bayesian Networks PGM 2022 Image De-Raining via Continual Learning CVPR 2021 Not all parameters are born equal: Attention is mostly what you need EMNLP 2021