Skip to content
conftrace_

weight disentanglement

6 papers

Explore in graph

Also known as

WD

Co-occurring keywords · papers mentioning both

task arithmetic (4) model editing (3) task vector (2) neural tangent kernel (2) zero-shot topic modeling (1) weight vector orthogonality (1) parameter fine-tuning (1) sparse fine-tuning (1) transformer language model (1) attention module (1)

Papers

Understanding and Enforcing Weight Disentanglement in Task Arithmetic CVPR 2026 Efficient Model Editing with Task-Localized Sparse Fine-tuning ICLR 2025 Fine-Tuning Attention Modules Only: Enhancing Weight Disentanglement in Task Arithmetic ICLR 2025 Mastering Task Arithmetic: τJp as a Key Indicator for Weight Disentanglement ICLR 2025 Disentangling Transformer Language Models as Superposed Topic Models EMNLP 2023 Task Arithmetic in the Tangent Space: Improved Editing of Pre-Trained Models NeurIPS 2023