通过进化策略拓展模型合并的参数空间，提升多任务性能。
Model Merging to Evolution: Parameter Space Exploration for Expert Models
- 将模型合并与进化结合，以合并结果为起点探索参数空间。
- 在单任务和多任务基准上表现媲美先进合并方法。
- 适合追求高性能合并模型的研究者和工程师。
模型合并通过整合多个专家模型的能力，在无需额外训练的情况下构建适用于多任务的强大模型，从而降低计算资源需求。然而，现有方法局限于专家模型的凸组合空间，无法探索该空间外的高性能区域。本文提出MERGEvolve框架，将模型合并与进化统一于进化策略中，将合并模型作为参数空间进化的初始点。合并阶段，专家模型作为确定性源构建强初始点；进化阶段则利用随机噪声探索参数空间。理论分析表明，MERGEvolve可探索凸组合空间之外的区域。在单任务和多任务基准上的大量实验显示，MERGEvolve始终达到与先进模型合并基线相当的性能。消融研究确认，高质量初始点对高效探索参数空间至关重要。
原文摘要 · Abstract (English)
Model merging integrates the capabilities of multiple expert models to create strong models for multiple tasks without additional training, thereby reducing computational resource requirements. However, existing methods operate within the convex combination space of expert models, failing to explore high-performance regions outside this space. This paper proposes the MERGEvolve framework, which unifies model merging and evolution within an evolution strategy by treating the merged model as the initialization for evolutionary exploration of the parameter space. During the merging phase, expert models act as deterministic sources to build a strong initial point. The evolution phase then explores the parameter space using random noise. Theoretical analysis shows that MERGEvolve explores regions outside the convex combination space. Extensive experiments on single-task and multi-task benchmarks demonstrate that MERGEvolve consistently achieves performance competitive with advanced model merging baselines. Ablation studies confirm that a high-quality initial point is critical for efficient exploration of the parameter space.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。