CoBa：用于大型语言模型多任务微调的收敛平衡器
CoBa: Convergence Balancer for Multitask Finetuning of Large Language Models
AI总结：
本文提出CoBa多任务学习方法，通过RCS、ACS和DF动态调整任务权重，以低开销平衡LLM各任务收敛，并在三个数据集上较次优基线最高提升13%性能。
摘要：
多任务学习（MTL）通过提供一个在各项任务上具有改进性能和泛化能力的单一模型，为大型语言模型（LLM）的微调带来益处，成为为每项任务开发独立模型的一种资源高效替代方案。然而，现有的面向LLM的MTL策略往往存在不足：要么计算开销巨大，要么无法确保各项任务同时收敛。本文提出了CoBa，这是一种旨在以最小计算开销有效管理任务收敛平衡的新型MTL方法。CoBa利用相对收敛分数（RCS）、绝对收敛分数（ACS）和发散因子（DF），在训练过程中动态调整任务权重，确保所有任务的验证损失以均匀的节奏向收敛推进，同时缓解单个任务发散的问题。我们在三个不同数据集上的实验结果表明，该方法不仅促进了任务收敛的均衡，而且相对于次优基线将LLM的性能提升了最高13%。代码已在 https://github.com/codefuse-ai/MFTCoder 开源。
英文摘要：
Multi-task learning (MTL) benefits the fine-tuning of large language models (LLMs) by providing a single model with improved performance and generalization ability across tasks, presenting a resource-efficient alternative to developing separate models for each task. Yet, existing MTL strategies for LLMs often fall short by either being computationally intensive or failing to ensure simultaneous task convergence. This paper presents CoBa, a new MTL approach designed to effectively manage task convergence balance with minimal computational overhead. Utilizing Relative Convergence Scores (RCS), Absolute Convergence Scores (ACS), and a Divergence Factor (DF), CoBa dynamically adjusts task weights during the training process, ensuring that the validation loss of all tasks progress towards convergence at an even pace while mitigating the issue of individual task divergence. The results of our experiments involving three disparate datasets underscore that this approach not only fosters equilibrium in task convergence but enhances the LLMs' performance by up to 13% relative to the second-best baselines. Code is open-sourced at https://github.com/codefuse-ai/MFTCoder.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
