小模型也能精准引用来源，提升生成内容可信度。
Even Small Reasoners Should Quote Their Sources: Introducing the Pleias-RAG Model Family
- 基于合成数据训练，原生支持引用和事实锚定。
- 350M与1B参数模型在多个基准上超越同类小模型。
- 适合资源受限环境，特别适用于多语种可靠问答场景。
我们推出新一代小型推理模型Pleias-RAG系列，专为RAG、检索与源文本摘要设计。Pleias-RAG-350m与Pleias-RAG-1B在大型合成数据集上进行中等规模训练，模拟从Common Corpus中检索多种语言开放源的场景。模型原生支持引用和事实锚定，包含字面引用，并整合查询路由、重写与源文档重排序等功能。在标准化RAG基准（HotPotQA、2wiki）上，这两款模型在参数量低于40亿的小模型中表现领先，且与Qwen-2.5-7B、Llama-3.1-8B、Gemma-3-4B等主流大模型性能相当。它们是目前唯一能在主流欧洲语言中保持稳定RAG表现，并确保陈述系统性引用依据的小模型。凭借小巧体积与部署友好性，以及设计上更高的事实准确性，该系列模型为生成式AI开辟了新应用空间。
原文摘要 · Abstract (English)
We introduce a new generation of small reasoning models for RAG, search, and source summarization. Pleias-RAG-350m and Pleias-RAG-1B are mid-trained on a large synthetic dataset emulating the retrieval of a wide variety of multilingual open sources from the Common Corpus. They provide native support for citation and grounding with literal quotes and reintegrate multiple features associated with RAG workflows, such as query routing, query reformulation, and source reranking. Pleias-RAG-350m and Pleias-RAG-1B outperform SLMs below 4 billion parameters on standardized RAG benchmarks (HotPotQA, 2wiki) and are competitive with popular larger models, including Qwen-2.5-7B, Llama-3.1-8B, and Gemma-3-4B. They are the only SLMs to date maintaining consistent RAG performance across leading European languages and ensuring systematic reference grounding for statements. Due to their size and ease of deployment on constrained infrastructure and higher factuality by design, the models unlock a range of new use cases for generative AI.
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。