Abstract
The emergence of large language models (LLMs) has significantly transformed natural language processing (NLP), enabling more generalized models to perform various tasks with minimal training. However, traditional sentiment analysis methods, which focus on individual tasks such as sentiment classification or aspect-based analysis, are not practical for real-world applications that usually require handling multiple tasks. While offering flexibility, LLMs in sentiment-specific tasks often fall short of the required accuracy. Techniques like fine-tuning and evolutionary model merging help integrate models into a unified framework, which can improve the learning performance while reducing computational costs. The use of task meta-data and curriculum learning to optimize learning processes remains underexplored, while sentiment analysis is a critical task in NLP that requires high accuracy and scalability across multiple subtasks. In this study, we propose a hybrid learning model called Multi-stage Evolutionary Model Merging with Meta data driven Curriculum Learning (MEM-MCL), to enhance the sentiment analysis in large language modeling. In particular, expert models are created through instruction tuning for specific sentiment tasks and then merged using evolutionary algorithms to form a unified model. The merging process is optimized with weak data to enhance performance across tasks. The curriculum learning is incorporated to provide a learning sequence based on task difficulty, improving knowledge extraction from LLMs. Experiment results demonstrate that the proposed MEM-MCL model outperforms conventional LLMs in a majority of sentiment analysis tasks, achieving superior results across various subtasks.
Abstract (translated)
大型语言模型（LLMs）的出现显著地改变了自然语言处理（NLP），使得更通用的模型能够在极少训练的情况下执行各种任务。然而，传统的侧重于情感分类或基于方面分析等单一任务的情感分析方法，在通常需要处理多项任务的实际应用中并不实用。虽然提供了灵活性，但在特定的情感任务中，大型语言模型往往达不到所需的准确度。通过微调和进化模型融合技术可以帮助将这些模型整合到统一的框架中，这可以提高学习性能同时减少计算成本。使用任务元数据和课程学习来优化学习过程尚处于未充分探索阶段，而情感分析作为NLP中的一个关键任务，需要在多项子任务上实现高准确性和可扩展性。 在此研究中，我们提出了一种混合学习模型——多阶段进化模型融合与基于元数据驱动的课程学习（MEM-MCL），以增强大型语言模型的情感分析能力。具体来说，通过指令微调为特定情感任务创建专家模型，并使用进化算法将这些模型合并成一个统一模型。在合并过程中利用弱标签数据来优化跨任务性能。此外，该模型还融入了基于任务难度的学习序列的课程学习，以提高从大型语言模型中提取知识的能力。 实验结果表明，所提出的MEM-MCL模型在多数情感分析任务上优于传统的LLMs，并且在各种子任务上都取得了更好的效果。
URL
https://arxiv.org/abs/2601.06780
