Cover Image

Beyond Task Vectors: Selective Task Arithmetic Based on Importance Metrics

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Bowen, Tian, Songning, Lai, Jiemin, Wu, Zhihao, Shuai, Shiming, Ge, Yutao, Yue
Format:	Preprint
Published:	2024
Subjects:	Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Text2Weight: Bridging Natural Language and Neural Network Weight Spaces
by: Tian, Bowen, et al.
Published: (2025)

Wolf2Pack: The AutoFusion Framework for Dynamic Parameter Fusion
by: Tian, Bowen, et al.
Published: (2024)

You Only Train Once: A Flexible Training Framework for Code Vulnerability Detection Driven by Vul-Vector
by: Tian, Bowen, et al.
Published: (2025)

Mitigating Spurious Background Bias in Multimedia Recognition with Disentangled Concept Bottlenecks
by: Huang, Gaoxiang, et al.
Published: (2025)

On Fairness of Task Arithmetic: The Role of Task Vectors
by: Naganuma, Hiroki, et al.
Published: (2025)

Task Vector Bases: A Unified and Scalable Framework for Compressed Task Arithmetic
by: Zeng, Siqi, et al.
Published: (2025)

IMTS is Worth Time $\times$ Channel Patches: Visual Masked Autoencoders for Irregular Multivariate Time Series Prediction
by: Hu, Zhangyi, et al.
Published: (2025)

Escaping Optimization Stagnation: Taking Steps Beyond Task Arithmetic via Difference Vectors
by: Wang, Jinping, et al.
Published: (2025)

Bus-Conditioned Zero-Shot Trajectory Generation via Task Arithmetic
by: Liu, Shuai, et al.
Published: (2026)

Provable In-Context Vector Arithmetic via Retrieving Task Concepts
by: Bu, Dake, et al.
Published: (2025)

Beyond the Node: Clade-level Selection for Efficient MCTS in Automatic Heuristic Design
by: Lai, Kezhao, et al.
Published: (2026)

Beyond Distribution Sharpening: The Importance of Task Rewards
by: Mittal, Sarthak, et al.
Published: (2026)

PEPL: Precision-Enhanced Pseudo-Labeling for Fine-Grained Image Classification in Semi-Supervised Learning
by: Tian, Bowen, et al.
Published: (2024)

Backdoor Vectors: a Task Arithmetic View on Backdoor Attacks and Defenses
by: Pawlak, Stanisław, et al.
Published: (2025)

Leveraging Submodule Linearity Enhances Task Arithmetic Performance in LLMs
by: Dai, Rui, et al.
Published: (2025)

FTS: A Framework to Find a Faithful TimeSieve
by: Lai, Songning, et al.
Published: (2024)

Layer-Aware Task Arithmetic: Disentangling Task-Specific and Instruction-Following Knowledge
by: Chen, Yan-Lun, et al.
Published: (2025)

Language Models Do Hard Arithmetic Tasks Easily and Hardly Do Easy Arithmetic Tasks
by: Gambardella, Andrew, et al.
Published: (2024)

On Task Vectors and Gradients
by: Zhou, Luca, et al.
Published: (2025)

Task Arithmetic Through The Lens Of One-Shot Federated Learning
by: Tao, Zhixu Silvia, et al.
Published: (2024)

Per-parameter Task Arithmetic for Unlearning in Large Language Models
by: Cai, Chengyi, et al.
Published: (2026)

DRIVE: Dual-Robustness via Information Variability and Entropic Consistency in Source-Free Unsupervised Domain Adaptation
by: Xiao, Ruiqiang, et al.
Published: (2024)

Maintaining Informative Coherence: Migrating Hallucinations in Large Language Models via Absorbing Markov Chains
by: Wu, Jiemin, et al.
Published: (2024)

Task Vector in TTS: Toward Emotionally Expressive Dialectal Speech Synthesis
by: Feng, Pengchao, et al.
Published: (2025)

Variational Task Vector Composition
by: Zhang, Boyuan, et al.
Published: (2025)

On the Importance of Task Complexity in Evaluating LLM-Based Multi-Agent Systems
by: Tang, Bohan, et al.
Published: (2025)

SECURE: Stable Early Collision Understanding via Robust Embeddings in Autonomous Driving
by: Wang, Wenjing, et al.
Published: (2026)

TADS: Task-Aware Data Selection for Multi-Task Multimodal Pre-Training
by: Cheng, Guanjie, et al.
Published: (2026)

When is Task Vector Provably Effective for Model Editing? A Generalization Analysis of Nonlinear Transformers
by: Li, Hongkang, et al.
Published: (2025)

Selective Task Group Updates for Multi-Task Optimization
by: Jeong, Wooseong, et al.
Published: (2025)

Decomposing Task Vectors for Refined Model Editing
by: Damirchi, Hamed, et al.
Published: (2025)

Beyond Task Performance: A Metric-Based Analysis of Sequential Cooperation in Heterogeneous Multi-Agent Destructive Foraging
by: Barrionuevo, Alejandro Mendoza, et al.
Published: (2026)

Task Vector Geometry Underlies Dual Modes of Task Inference in Transformers
by: Yan, Hao, et al.
Published: (2026)

Fine-Tuning Attention Modules Only: Enhancing Weight Disentanglement in Task Arithmetic
by: Jin, Ruochen, et al.
Published: (2024)

Investigating Task Arithmetic for Zero-Shot Information Retrieval
by: Braga, Marco, et al.
Published: (2025)

Task Vector Quantization for Memory-Efficient Model Merging
by: Kim, Youngeun, et al.
Published: (2025)

Task Vectors in In-Context Learning: Emergence, Formation, and Benefit
by: Yang, Liu, et al.
Published: (2025)

TaskMet: Task-Driven Metric Learning for Model Learning
by: Bansal, Dishank, et al.
Published: (2023)

Principled Understanding of Generalization for Generative Transformer Models in Arithmetic Reasoning Tasks
by: Xu, Xingcheng, et al.
Published: (2024)

Explicitly Encoding Structural Symmetry is Key to Length Generalization in Arithmetic Tasks
by: Sabbaghi, Mahdi, et al.
Published: (2024)