通过分层合并多模型，显著提升大模型生成一致性。
Harmonizing Diverse Models: A Layer-wise Merging Strategy for Consistent Generation
- 基于中间层激活值设计一致性权重，分层融合多个专用模型。
- 在测试中使输出相似度提升约47.5%，优于基线模型。
- 适合需要高可靠性的工业级检索增强生成系统使用。
检索增强生成（RAG）系统利用大语言模型（LLM）生成基于检索上下文的准确且可靠的回答。然而，大语言模型对语义等价输入常产生不一致输出，这一问题因一致性训练数据稀缺及现有微调技术在提升输出一致性方面的局限而加剧。本文提出一种新方法，结合系统化合成数据生成、三元组损失以优化嵌入表示，以及一种新颖的分层模型合并策略。通过从中间层激活中获取的一致性感知权重，该方法有效整合了多个专用模型的知识。实验结果表明，所合并模型显著提升了输出一致性，在响应相似度上相比基线模型提升约47.5%，为工业级RAG系统的可靠性提升提供了实用解决方案。
原文摘要 · Abstract (English)
Retrieval-Augmented Generation (RAG) systems leverage Large Language Models (LLMs) to generate accurate and reliable responses that are grounded in retrieved context. However, LLMs often generate inconsistent outputs for semantically equivalent inputs, a problem compounded by the scarcity of consistency-focused training data and the limitations of current fine-tuning techniques in enhancing output consistency. We propose a new approach combining systematic synthetic data generation, triplet loss for better embeddings, and a novel layer-wise model merging approach. Using consistency-aware weights derived from intermediate layer activations, our method effectively integrates knowledge from specialized models. Experimental results how that our merged model significantly enhances output consistency, achieving a ~47.5\% improvement in response similarity over the baseline, thus offering a practical solution for increasing the reliability of an industrial RAG system.
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。