Cover Image

Preference-Aligned LoRA Merging: Preserving Subspace Coverage and Addressing Directional Anisotropy

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Jeong, Wooseong, Lee, Wonyoung, Yoon, Kuk-Jin
Format:	Preprint
Published:	2026
Subjects:	Computer Vision and Pattern Recognition Artificial Intelligence Machine Learning
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

Label-Free Cross-Task LoRA Merging with Null-Space Compression
by: Lee, Wonyoung, et al.
Published: (2026)

Resolving Token-Space Gradient Conflicts: Token Space Manipulation for Transformer-Based Multi-Task Learning
by: Jeong, Wooseong, et al.
Published: (2025)

Synchronizing Task Behavior: Aligning Multiple Tasks during Test-Time Training
by: Jeong, Wooseong, et al.
Published: (2025)

Interaction-Merged Motion Planning: Effectively Leveraging Diverse Motion Datasets for Robust Planning
by: Lee, Giwon, et al.
Published: (2025)

Shared LoRA Subspaces for almost Strict Continual Learning
by: Kaushik, Prakhar, et al.
Published: (2026)

Quantifying Task Priority for Multi-Task Optimization
by: Jeong, Wooseong, et al.
Published: (2024)

LoRA.rar: Learning to Merge LoRAs via Hypernetworks for Subject-Style Conditioned Image Generation
by: Shenaj, Donald, et al.
Published: (2024)

Rethinking Inter-LoRA Orthogonality in Adapter Merging: Insights from Orthogonal Monte Carlo Dropout
by: Zhang, Andi, et al.
Published: (2025)

MapReduce LoRA: Advancing the Pareto Front in Multi-Preference Optimization for Generative Models
by: Chen, Chieh-Yun, et al.
Published: (2025)

RecycleLoRA: Rank-Revealing QR-Based Dual-LoRA Subspace Adaptation for Domain Generalized Semantic Segmentation
by: Cho, Chanseul, et al.
Published: (2026)

Evolutionary Negative Module Pruning for Better LoRA Merging
by: Cao, Anda, et al.
Published: (2026)

Ortho-Hydra: Orthogonalized Experts for DiT LoRA
by: Ji, Seunghyun
Published: (2026)

FACL-Attack: Frequency-Aware Contrastive Learning for Transferable Adversarial Attacks
by: Yang, Hunmin, et al.
Published: (2024)

FM-LoRA: Factorized Low-Rank Meta-Prompting for Continual Learning
by: Yu, Xiaobing, et al.
Published: (2025)

Continual Diffusion: Continual Customization of Text-to-Image Diffusion with C-LoRA
by: Smith, James Seale, et al.
Published: (2023)

Vertical LoRA: Dense Expectation-Maximization Interpretation of Transformers
by: Fu, Zhuolin
Published: (2024)

LoRA in LoRA: Towards Parameter-Efficient Architecture Expansion for Continual Visual Instruction Tuning
by: Che, Chang, et al.
Published: (2025)

Dual-Space Augmented Intrinsic-LoRA for Wind Turbine Segmentation
by: Singhal, Shubh, et al.
Published: (2024)

Selective LoRA for Visual Tokens and Attention Heads
by: Luo, Tiange, et al.
Published: (2025)

Multi LoRA Meets Vision: Merging multiple adapters to create a multi task model
by: Kesim, Ege, et al.
Published: (2024)

LoRA Subtraction for Drift-Resistant Space in Exemplar-Free Continual Learning
by: Liu, Xuan, et al.
Published: (2025)

Block-wise LoRA: Revisiting Fine-grained LoRA for Effective Personalization and Stylization in Text-to-Image Generation
by: Li, Likun, et al.
Published: (2024)

Prompt-Driven Contrastive Learning for Transferable Adversarial Attacks
by: Yang, Hunmin, et al.
Published: (2024)

Multi-LoRA Composition for Image Generation
by: Zhong, Ming, et al.
Published: (2024)

Empower Vision Applications with LoRA LMM
by: Mi, Liang, et al.
Published: (2024)

Adaptive Shared Experts with LoRA-Based Mixture of Experts for Multi-Task Learning
by: Yang, Minghao, et al.
Published: (2025)

DAC-LoRA: Dynamic Adversarial Curriculum for Efficient and Robust Few-Shot Adaptation
by: Umrajkar, Ved
Published: (2025)

Simple Drop-in LoRA Conditioning on Attention Layers Will Improve Your Diffusion Model
by: Choi, Joo Young, et al.
Published: (2024)

EMLoC: Emulator-based Memory-efficient Fine-tuning with LoRA Correction
by: Lin, Hsi-Che, et al.
Published: (2025)

Not All Layers Are Created Equal: Adaptive LoRA Ranks for Personalized Image Generation
by: Shenaj, Donald, et al.
Published: (2026)

CoLoRSMamba: Conditional LoRA-Steered Mamba for Supervised Multimodal Violence Detection
by: Senadeera, Damith Chamalke, et al.
Published: (2026)

Contrastive Regularization over LoRA for Multimodal Biomedical Image Incremental Learning
by: Zhang, Haojie, et al.
Published: (2025)

Multi-agent Long-term 3D Human Pose Forecasting via Interaction-aware Trajectory Conditioning
by: Jeong, Jaewoo, et al.
Published: (2024)

Subspace-Boosted Model Merging
by: Skorobogat, Ronald, et al.
Published: (2025)

In-Context Meta LoRA Generation
by: Shao, Yihua, et al.
Published: (2025)

Robust Adverse Weather Removal via Spectral-based Spatial Grouping
by: Jeong, Yuhwan, et al.
Published: (2025)

Frugal Federated Learning for Violence Detection: A Comparison of LoRA-Tuned VLMs and Personalized CNNs
by: Thuau, Sébastien, et al.
Published: (2025)

Improving Black-Box Generative Attacks via Generator Semantic Consistency
by: Jeong, Jongoh, et al.
Published: (2025)

How LoRA Remembers? A Parametric Memory Law for LLM Finetuning
by: Xu, Ziwen, et al.
Published: (2026)

DyME: Dynamic Multi-Concept Erasure in Diffusion Models with Bi-Level Orthogonal LoRA Adaptation
by: Liu, Jiaqi, et al.
Published: (2025)