SOURCE-LINKED INTELLIGENCE

CoMerge: Conflict-Driven Preference Optimization for Multi-Task Model Merging

arXiv · AI, language, vision and robotics · article · Sep 2, 2026 · UTC

Model merging provides an efficient paradigm for constructing multi-task large language models (LLMs) without full model retraining, yet it remains challenged by parameter interference. While existing methods aim to preserve the capabilities of individual expert models and mitigate interference, they generally do not directly learn from the potentially degraded behaviors exposed by naive merging. In this paper, we propose a conflict-driven preference optimization framework for model merging (CoMerge), which reformulates model merging as a preference optimization problem. The approach utilizes

Read original source ↗ Open in workspace

recordType

paper

region

Global