CAT Merging: A Training-Free Approach for Resolving Conflicts in Model Merging

2025-05-11 · Wenju Sun, Qingyong Li, Yangli-ao Geng, Boyang Li

Multi-task model merging offers a promising paradigm for integrating multiple expert models into a unified model without additional training. Existing state-of-the-art techniques, such as Task Arithmetic and its variants, merge models by accumulating task vectors -- the parameter differences between pretrained and finetuned models. However, task vector accumulation is often hindered by knowledge conflicts, leading to performance degradation. To address this challenge, we propose Conflict-Aware Task Merging (CAT Merging), a novel training-free framework that selectively trims conflict-prone components from the task vectors. CAT Merging introduces several parameter-specific strategies, including projection for linear weights and masking for scaling and shifting parameters in normalization layers. Extensive experiments on vision, language, and vision-language tasks demonstrate that CAT Merging effectively suppresses knowledge conflicts, achieving average accuracy improvements of up to 2.5% (ViT-B/32) and 2.0% (ViT-L/14) over state-of-the-art methods.

📄 PDF Abstract BibTeX arXiv:2505.06977

Code (0)

등록된 구현이 없습니다.

Tasks

Task Arithmetic

When Shared Knowledge Hurts: Spectral Over-Accumulation in Model Merging

2026-02-05 · Yayuan Li, Ze Peng, Jian Zhang, Jintao Guo 외 arxiv

Model merging combines multiple fine-tuned models into a single model by adding their weight updates, providing a lightweight alternative to retraining. Existing methods primarily target resolving conflicts between task …

TIES-Merging: Resolving Interference When Merging Models

2023-06-02 · NeurIPS 2023 11 · Prateek Yadav, Derek Tam, Leshem Choshen, Colin Raffel 외

Transfer learning - i.e., further fine-tuning a pre-trained model on a downstream task - can confer significant advantages, including improved downstream performance, faster convergence, and better sample efficiency. The…

Transfer Learning

LED-Merging: Mitigating Safety-Utility Conflicts in Model Merging with Location-Election-Disjoint

2025-02-24 · Qianli Ma, Dongrui Liu, Qian Chen, Linfeng Zhang 외

Fine-tuning pre-trained Large Language Models (LLMs) for specialized tasks incurs substantial computational and data costs. While model merging offers a training-free solution to integrate multiple task-specific models, …

GSM8K

Bridging Domains through Subspace-Aware Model Merging

2026-03-06 · Levy Chaves, Chao Zhou, Rebekka Burkholz, Eduardo Valle 외 arxiv

Model merging integrates multiple task-specific models into a single consolidated one. Recent research has made progress in improving merging performance for in-distribution or multi-task scenarios, but domain generaliza…

Domain Generalization

Sparsity Curse: Understanding RLVR Model Parameter Space from Model Merging

2026-06-16 · Chenrui Wu, Zexi Li, Jiajun Bu, Jiangchuan Liu 외 arxiv

Reinforcement Learning with Verifiable Reward (RLVR) has emerged as a powerful post-training paradigm that surpasses Supervised Fine-Tuning (SFT) in eliciting reasoning intelligence and resisting catastrophic forgetting.…

Reinforcement Learning