Image de couverture de livre

Transporting Task Vectors across Different Architectures without Training

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Rinaldi, Filippo, Panariello, Aniello, Salici, Giacomo, Porrello, Angelo, Calderara, Simone
Format:	Preprint
Publié:	2026
Sujets:	Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Gradient-Sign Masking for Task Vector Transport Across Pre-Trained Models
par: Rinaldi, Filippo, et autres
Publié: (2025)

CLIP with Generative Latent Replay: a Strong Baseline for Incremental Learning
par: Frascaroli, Emanuele, et autres
Publié: (2024)

Modular Embedding Recomposition for Incremental Learning
par: Panariello, Aniello, et autres
Publié: (2025)

Is Multiple Object Tracking a Matter of Specialization?
par: Mancusi, Gianluca, et autres
Publié: (2024)

Trajectory Forecasting through Low-Rank Adaptation of Discrete Latent Codes
par: Benaglia, Riccardo, et autres
Publié: (2024)

How to Train Your Metamorphic Deep Neural Network
par: Sommariva, Thomas, et autres
Publié: (2025)

May the Forgetting Be with You: Alternate Replay for Learning with Noisy Labels
par: Millunzi, Monica, et autres
Publié: (2024)

Accurate and Efficient Low-Rank Model Merging in Core Space
par: Panariello, Aniello, et autres
Publié: (2025)

Mask and Compress: Efficient Skeleton-based Action Recognition in Continual Learning
par: Mosconi, Matteo, et autres
Publié: (2024)

Monocular Per-Object Distance Estimation with Masked Object Modeling
par: Panariello, Aniello, et autres
Publié: (2024)

Zero-Shot Synthetic-to-Real Handwritten Text Recognition via Task Analogies
par: Garrido-Munoz, Carlos, et autres
Publié: (2026)

Distilling Linearized Behavior into Non-Linear Fine-Tuning for Effective Task Arithmetic
par: Sommariva, Thomas, et autres
Publié: (2026)

An Attention-based Representation Distillation Baseline for Multi-Label Continual Learning
par: Menabue, Martin, et autres
Publié: (2024)

DitHub: A Modular Framework for Incremental Open-Vocabulary Object Detection
par: Cappellino, Chiara, et autres
Publié: (2025)

Merging without Forgetting: Continual Fusion of Task-Specific Models via Optimal Transport
par: Pan, Zecheng, et autres
Publié: (2025)

MultiLink: Multi-class Structure Recovery via Agglomerative Clustering and Model Selection
par: Magri, Luca, et autres
Publié: (2025)

Video Reasoning without Training
par: Sridhar, Deepak, et autres
Publié: (2025)

PIF: Anomaly detection via preference embedding
par: Leveni, Filippo, et autres
Publié: (2025)

Hashing for Structure-based Anomaly Detection
par: Leveni, Filippo, et autres
Publié: (2025)

Preference Isolation Forest for Structure-based Anomaly Detection
par: Leveni, Filippo, et autres
Publié: (2025)

Simplifying Multi-Task Architectures Through Task-Specific Normalization
par: Suteu, Mihai, et autres
Publié: (2025)

Update Your Transformer to the Latest Release: Re-Basin of Task Vectors
par: Rinaldi, Filippo, et autres
Publié: (2025)

Knowledge Composition using Task Vectors with Learned Anisotropic Scaling
par: Zhang, Frederic Z., et autres
Publié: (2024)

Visual RAG: Expanding MLLM visual knowledge without fine-tuning
par: Bonomo, Mirco, et autres
Publié: (2025)

An Architecture Built for Federated Learning: Addressing Data Heterogeneity through Adaptive Normalization-Free Feature Recalibration
par: Siomos, Vasilis, et autres
Publié: (2024)

Synchronizing Task Behavior: Aligning Multiple Tasks during Test-Time Training
par: Jeong, Wooseong, et autres
Publié: (2025)

A Second-Order Perspective on Model Compositionality and Incremental Learning
par: Porrello, Angelo, et autres
Publié: (2024)

Self-Labeling the Job Shop Scheduling Problem
par: Corsini, Andrea, et autres
Publié: (2024)

Structure-based Anomaly Detection and Clustering
par: Leveni, Filippo
Publié: (2025)

Transferring Visual Explainability of Self-Explaining Models to Prediction-Only Models without Additional Training
par: Yoshikawa, Yuya, et autres
Publié: (2025)

VideoGuide: Improving Video Diffusion Models without Training Through a Teacher's Guide
par: Lee, Dohun, et autres
Publié: (2024)

Trained Models Tell Us How to Make Them Robust to Spurious Correlation without Group Annotation
par: Ghaznavi, Mahdi, et autres
Publié: (2024)

Curriculum Multi-Task Self-Supervision Improves Lightweight Architectures for Onboard Satellite Hyperspectral Image Segmentation
par: Carlesso, Hugo, et autres
Publié: (2025)

Proactive Gradient Conflict Mitigation in Multi-Task Learning: A Sparse Training Perspective
par: Zhang, Zhi, et autres
Publié: (2024)

Generative AI-based Pipeline Architecture for Increasing Training Efficiency in Intelligent Weed Control Systems
par: Modak, Sourav, et autres
Publié: (2024)

ZipIt! Merging Models from Different Tasks without Training
par: Stoica, George, et autres
Publié: (2023)

Multimodal Task Vectors Enable Many-Shot Multimodal In-Context Learning
par: Huang, Brandon, et autres
Publié: (2024)

Convolutional Set Transformer
par: Chinello, Federico, et autres
Publié: (2025)

Predicting Depression and Anxiety Risk in Dutch Neighborhoods from Street-View Images
par: Khodorivsko, Nin, et autres
Publié: (2024)

Balancing Accuracy and Training Time in Federated Learning for Violence Detection in Surveillance Videos: A Study of Neural Network Architectures
par: Quentin, Pajon, et autres
Publié: (2023)