Enregistré dans:
| Auteurs principaux: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | https://arxiv.org/abs/2504.18225 |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909594143424512 |
|---|---|
| author | Langlais, Pierre-Carl Chizhov, Pavel Nee, Mattia Hinostroza, Carlos Rosas Delsart, Matthieu Girard, Irène Hicheur, Othman Stasenko, Anastasia Yamshchikov, Ivan P. |
| author_facet | Langlais, Pierre-Carl Chizhov, Pavel Nee, Mattia Hinostroza, Carlos Rosas Delsart, Matthieu Girard, Irène Hicheur, Othman Stasenko, Anastasia Yamshchikov, Ivan P. |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_18225 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Even Small Reasoners Should Quote Their Sources: Introducing the Pleias-RAG Model Family Langlais, Pierre-Carl Chizhov, Pavel Nee, Mattia Hinostroza, Carlos Rosas Delsart, Matthieu Girard, Irène Hicheur, Othman Stasenko, Anastasia Yamshchikov, Ivan P. Computation and Language 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. |
| title | Even Small Reasoners Should Quote Their Sources: Introducing the Pleias-RAG Model Family |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2504.18225 |