Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2501.13880 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912201523068928 |
|---|---|
| author | Kuratomi, Gustavo Pirozelli, Paulo Cozman, Fabio G. Peres, Sarajane M. |
| author_facet | Kuratomi, Gustavo Pirozelli, Paulo Cozman, Fabio G. Peres, Sarajane M. |
| contents | Although large language models (LLMs) demonstrate strong text generation capabilities, they struggle in scenarios requiring access to structured knowledge bases or specific documents, limiting their effectiveness in knowledge-intensive tasks. To address this limitation, retrieval-augmented generation (RAG) models have been developed, enabling generative models to incorporate relevant document fragments into their inputs. In this paper, we design and evaluate a RAG-based virtual assistant specifically tailored for the University of São Paulo. Our system architecture comprises two key modules: a retriever and a generative model. We experiment with different types of models for both components, adjusting hyperparameters such as chunk size and the number of retrieved documents. Our optimal retriever model achieves a Top-5 accuracy of 30%, while our most effective generative model scores 22.04\% against ground truth answers. Notably, when the correct document chunks are supplied to the LLMs, accuracy significantly improves to 54.02%, an increase of over 30 percentage points. Conversely, without contextual input, performance declines to 13.68%. These findings highlight the critical role of database access in enhancing LLM performance. They also reveal the limitations of current semantic search methods in accurately identifying relevant documents and underscore the ongoing challenges LLMs face in generating precise responses. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_13880 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | A RAG-Based Institutional Assistant Kuratomi, Gustavo Pirozelli, Paulo Cozman, Fabio G. Peres, Sarajane M. Computation and Language Although large language models (LLMs) demonstrate strong text generation capabilities, they struggle in scenarios requiring access to structured knowledge bases or specific documents, limiting their effectiveness in knowledge-intensive tasks. To address this limitation, retrieval-augmented generation (RAG) models have been developed, enabling generative models to incorporate relevant document fragments into their inputs. In this paper, we design and evaluate a RAG-based virtual assistant specifically tailored for the University of São Paulo. Our system architecture comprises two key modules: a retriever and a generative model. We experiment with different types of models for both components, adjusting hyperparameters such as chunk size and the number of retrieved documents. Our optimal retriever model achieves a Top-5 accuracy of 30%, while our most effective generative model scores 22.04\% against ground truth answers. Notably, when the correct document chunks are supplied to the LLMs, accuracy significantly improves to 54.02%, an increase of over 30 percentage points. Conversely, without contextual input, performance declines to 13.68%. These findings highlight the critical role of database access in enhancing LLM performance. They also reveal the limitations of current semantic search methods in accurately identifying relevant documents and underscore the ongoing challenges LLMs face in generating precise responses. |
| title | A RAG-Based Institutional Assistant |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2501.13880 |