一种基于RAG的强化学习课程学习方法：用于多模态问答
A Curriculum Learning Approach to Reinforcement Learning: Leveraging RAG for Multimodal Question Answering
AI总结：
本文提出了一种结合课程学习与强化学习的方法，用于多模态问答任务，通过检索增强生成系统在挑战中取得优异成绩。
摘要：
本文描述了Dianping-Trust-Safety团队在META CRAG-MM挑战中的解决方案。该挑战要求构建一个能够处理多模态多轮问答的综合性检索增强生成系统。比赛包含三个任务：（1）使用从基于图像的模拟知识图中检索到的结构化数据来回答问题；（2）从知识图和网络搜索结果中综合信息；（3）处理需要上下文理解和多来源信息聚合的多轮对话。对于任务1，我们的解决方案基于视觉大语言模型，通过监督微调增强，利用从GPT-4.1中提炼的知识。我们进一步应用了课程学习策略来指导强化学习，从而提高了回答的准确性并减少了幻觉。对于任务2和任务3，我们还利用了网络搜索API来引入外部知识，使系统能够更好地处理复杂查询和多轮对话。我们的方法在任务1中取得了第一名，领先幅度达52.38%，在任务3中获得第三名，证明了我们在训练流程中将课程学习与强化学习相结合的有效性。
英文摘要：
This paper describes the solutions of the Dianping-Trust-Safety team for the META CRAG-MM challenge. The challenge requires building a comprehensive retrieval-augmented generation system capable for multi-modal multi-turn question answering. The competition consists of three tasks: (1) answering questions using structured data retrieved from an image-based mock knowledge graph, (2) synthesizing information from both knowledge graphs and web search results, and (3) handling multi-turn conversations that require context understanding and information aggregation from multiple sources. For Task 1, our solution is based on the vision large language model, enhanced by supervised fine-tuning with knowledge distilled from GPT-4.1. We further applied curriculum learning strategies to guide reinforcement learning, resulting in improved answer accuracy and reduced hallucination. For Task 2 and Task 3, we additionally leveraged web search APIs to incorporate external knowledge, enabling the system to better handle complex queries and multi-turn conversations. Our approach achieved 1st place in Task 1 with a significant lead of 52.38%, and 3rd place in Task 3, demonstrating the effectiveness of the integration of curriculum learning with reinforcement learning in our training pipeline.
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。
