Retrieval-Augmented Guidance for Inclusive Teaching in Higher Education
Resumo
To support inclusive teaching practices in higher education, this work presents a Retrieval-Augmented Generation system that provides educators with assistance supported by peer-reviewed literature. The system combines a curated corpus that is automatically expanded as new research is discovered, with a document ingestion procedure that uses a VLM to preserve document structure. We benchmarked seven embedding models and six open-weights LLMs using the RAGAS framework, and the best configuration achieved a document-level Recall@10 of 0.91 and a Faithfulness score of 0.92, suggesting that RAG is a promising approach for supporting educators in adopting inclusive practices for teaching students with disabilities.
Palavras-chave:
Retrieval-Augmented Generation, Inclusive Higher Education, Educational Assistants
Referências
dos Santos, A. B., Santos, R. E. d. S., and da Silva, C. M. R. (2025). Teachat: Promovendo o suporte educacional ao autismo com inteligência artificial conversacional. In Workshop sobre Educação em Computação (WEI), pages 1369–1379. SBC.
Douze, M., Guzhva, A., Deng, C., Johnson, J., Szilvasy, G., Mazaré, P.-E., Lomeli, M., Hosseini, L., and Jégou, H. (2026). The faiss library. IEEE Transactions on Big Data, 12(2):346–361.
Es, S., James, J., Anke, L. E., and Schockaert, S. (2024). Ragas: Automated evaluation of retrieval augmented generation. In Proceedings of the 18th conference of the european chapter of the association for computational linguistics: system demonstrations, pages 150–158.
Galfre, M., Curcio, V., Dardouillet, P., and Le Van, O. (2025). Vision-language models for sick leave certificates: Beyond ocr in real-world form understanding. In 2025 3rd International Conference on Foundation and Large Language Models (FLLM), pages 254–261. IEEE.
Grattafiori, A., Dubey, A., Jauhri, A., Pandey, A., Kadian, A., Al-Dahle, A., Letman, A., Mathur, A., Schelten, A., Vaughan, A., et al. (2024). The llama 3 herd of models. Technical report, Meta AI. arXiv:2407.21783.
Günther, M., Ong, J., Mohr, I., Abdessalem, A., Abel, T., Akram, M. K., Guzman, S., Mastrapas, G., Sturua, S., Wang, B., et al. (2023). Jina embeddings 2: 8192-token general-purpose text embeddings for long documents. arXiv preprint arXiv:2310.19923.
Hennekeuser, D., Vaziri, D. D., Golchinfar, D., Schreiber, D., and Stevens, G. (2025). Enlarged education–exploring the use of generative ai to support lecturing in higher education. International Journal of Artificial Intelligence in Education, 35(3):1096–1128.
Hurst, A., Lerer, A., Goucher, A. P., Perelman, A., Ramesh, A., Clark, A., Ostrow, A., Welihinda, A., Hayes, A., Radford, A., et al. (2024). Gpt-4o system card. Technical Report arXiv:2410.21276, OpenAI.
Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W.-t., Rocktäschel, T., et al. (2020). Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in neural information processing systems, 33:9459–9474.
Liu, A. H., Khandelwal, K., Subramanian, S., Jouault, V., Rastogi, A., Sadé, A., Jeffares, A., Jiang, A., Cahill, A., Gavaudan, A., et al. (2026). Ministral 3. arXiv preprint arXiv:2601.08584.
Lopez-Gazpio, I. (2025). Integrating large language models into accessible and inclusive education: access democratization and individualized learning enhancement supported by generative artificial intelligence. Information, 16(6):473.
Matos, F. D., Girotto Junior, G., de Medeiros Arnt, A., and Lippi, A. (2024). A proposal in brazil to use generative ai in education threatens quality and equity. Patterns, 5(7):101020.
Mistral AI (2025). Mistral-small-3.1-24b-instruct-2503 model card. [link]. Acesso em: 22 mai. 2026.
Nussbaum, Z., Morris, J. X., Duderstadt, B., and Mulyar, A. (2024). Nomic embed: Training a reproducible long context text embedder.
OpenAI (2025). gpt-oss-120b & gpt-oss-20b model card. Model card, OpenAI. arXiv:2508.10925.
Pan, F., Zhou, Q., Guo, W., and Yang, H. (2025). A survey on retrieval-augmented generation in applications of education and teaching. In 2025 7th International Conference on Computer Science and Technologies in Education (CSTE), pages 803–807. IEEE.
Qwen (2025). Qwen3.5-27b model card. [link]. Model card. Acesso em: 22 mai. 2026.
Qwen Team (2024). Qwen2.5 technical report. Technical report, Alibaba Group. arXiv:2412.15115.
Reis-Andersson, J. (2024). Ai generative in brazil's public schools: The teachers' perspective. International Conference on AI Research, 4:374–380.
Song, K., Tan, X., Qin, T., Lu, J., and Liu, T.-Y. (2020). Mpnet: Masked and permuted pre-training for language understanding. In Advances in Neural Information Processing Systems, volume 33, pages 16857–16867.
UNDESA (2019). Realizing the Sustainable Development Goals by, for and with Persons with Disabilities. Stylus Publishing, LLC.
UNDESA (2025). Accelerating the Realization of the Sustainable Development Goals by, for and with Persons with Disabilities. Stylus Publishing, LLC.
United Nations (2015). Transforming our world: the 2030 agenda for sustainable development. Acesso em: 22 maio 2026.
Wang, L., Yang, N., Huang, X., Jiao, B., Yang, L., Jiang, D., Majumder, R., and Wei, F. (2022). Text embeddings by weakly-supervised contrastive pre-training. arXiv preprint arXiv:2212.03533.
Wang, W., Wei, F., Dong, L., Bao, H., Yang, N., and Zhou, M. (2020). Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers. arXiv.
Xiao, S., Liu, Z., Zhang, P., and Muennighoff, N. (2023). C-pack: Packaged resources to advance general chinese embedding.
Yu, S., Tang, C., Xu, B., Cui, J., Ran, J., Yan, Y., Liu, Z., Wang, S., Han, X., Liu, Z., et al. (2025). Visrag: Vision-based retrieval-augmented generation on multi-modality documents. In International Conference on Learning Representations, volume 2025.
Douze, M., Guzhva, A., Deng, C., Johnson, J., Szilvasy, G., Mazaré, P.-E., Lomeli, M., Hosseini, L., and Jégou, H. (2026). The faiss library. IEEE Transactions on Big Data, 12(2):346–361.
Es, S., James, J., Anke, L. E., and Schockaert, S. (2024). Ragas: Automated evaluation of retrieval augmented generation. In Proceedings of the 18th conference of the european chapter of the association for computational linguistics: system demonstrations, pages 150–158.
Galfre, M., Curcio, V., Dardouillet, P., and Le Van, O. (2025). Vision-language models for sick leave certificates: Beyond ocr in real-world form understanding. In 2025 3rd International Conference on Foundation and Large Language Models (FLLM), pages 254–261. IEEE.
Grattafiori, A., Dubey, A., Jauhri, A., Pandey, A., Kadian, A., Al-Dahle, A., Letman, A., Mathur, A., Schelten, A., Vaughan, A., et al. (2024). The llama 3 herd of models. Technical report, Meta AI. arXiv:2407.21783.
Günther, M., Ong, J., Mohr, I., Abdessalem, A., Abel, T., Akram, M. K., Guzman, S., Mastrapas, G., Sturua, S., Wang, B., et al. (2023). Jina embeddings 2: 8192-token general-purpose text embeddings for long documents. arXiv preprint arXiv:2310.19923.
Hennekeuser, D., Vaziri, D. D., Golchinfar, D., Schreiber, D., and Stevens, G. (2025). Enlarged education–exploring the use of generative ai to support lecturing in higher education. International Journal of Artificial Intelligence in Education, 35(3):1096–1128.
Hurst, A., Lerer, A., Goucher, A. P., Perelman, A., Ramesh, A., Clark, A., Ostrow, A., Welihinda, A., Hayes, A., Radford, A., et al. (2024). Gpt-4o system card. Technical Report arXiv:2410.21276, OpenAI.
Lewis, P., Perez, E., Piktus, A., Petroni, F., Karpukhin, V., Goyal, N., Küttler, H., Lewis, M., Yih, W.-t., Rocktäschel, T., et al. (2020). Retrieval-augmented generation for knowledge-intensive nlp tasks. Advances in neural information processing systems, 33:9459–9474.
Liu, A. H., Khandelwal, K., Subramanian, S., Jouault, V., Rastogi, A., Sadé, A., Jeffares, A., Jiang, A., Cahill, A., Gavaudan, A., et al. (2026). Ministral 3. arXiv preprint arXiv:2601.08584.
Lopez-Gazpio, I. (2025). Integrating large language models into accessible and inclusive education: access democratization and individualized learning enhancement supported by generative artificial intelligence. Information, 16(6):473.
Matos, F. D., Girotto Junior, G., de Medeiros Arnt, A., and Lippi, A. (2024). A proposal in brazil to use generative ai in education threatens quality and equity. Patterns, 5(7):101020.
Mistral AI (2025). Mistral-small-3.1-24b-instruct-2503 model card. [link]. Acesso em: 22 mai. 2026.
Nussbaum, Z., Morris, J. X., Duderstadt, B., and Mulyar, A. (2024). Nomic embed: Training a reproducible long context text embedder.
OpenAI (2025). gpt-oss-120b & gpt-oss-20b model card. Model card, OpenAI. arXiv:2508.10925.
Pan, F., Zhou, Q., Guo, W., and Yang, H. (2025). A survey on retrieval-augmented generation in applications of education and teaching. In 2025 7th International Conference on Computer Science and Technologies in Education (CSTE), pages 803–807. IEEE.
Qwen (2025). Qwen3.5-27b model card. [link]. Model card. Acesso em: 22 mai. 2026.
Qwen Team (2024). Qwen2.5 technical report. Technical report, Alibaba Group. arXiv:2412.15115.
Reis-Andersson, J. (2024). Ai generative in brazil's public schools: The teachers' perspective. International Conference on AI Research, 4:374–380.
Song, K., Tan, X., Qin, T., Lu, J., and Liu, T.-Y. (2020). Mpnet: Masked and permuted pre-training for language understanding. In Advances in Neural Information Processing Systems, volume 33, pages 16857–16867.
UNDESA (2019). Realizing the Sustainable Development Goals by, for and with Persons with Disabilities. Stylus Publishing, LLC.
UNDESA (2025). Accelerating the Realization of the Sustainable Development Goals by, for and with Persons with Disabilities. Stylus Publishing, LLC.
United Nations (2015). Transforming our world: the 2030 agenda for sustainable development. Acesso em: 22 maio 2026.
Wang, L., Yang, N., Huang, X., Jiao, B., Yang, L., Jiang, D., Majumder, R., and Wei, F. (2022). Text embeddings by weakly-supervised contrastive pre-training. arXiv preprint arXiv:2212.03533.
Wang, W., Wei, F., Dong, L., Bao, H., Yang, N., and Zhou, M. (2020). Minilm: Deep self-attention distillation for task-agnostic compression of pre-trained transformers. arXiv.
Xiao, S., Liu, Z., Zhang, P., and Muennighoff, N. (2023). C-pack: Packaged resources to advance general chinese embedding.
Yu, S., Tang, C., Xu, B., Cui, J., Ran, J., Yan, Y., Liu, Z., Wang, S., Han, X., Liu, Z., et al. (2025). Visrag: Vision-based retrieval-augmented generation on multi-modality documents. In International Conference on Learning Representations, volume 2025.
Publicado
05/10/2026
Como Citar
HARADA, William Massami Costa; DE MELO, Tiago Eugênio; DA SILVA, Fábio Santos.
Retrieval-Augmented Guidance for Inclusive Teaching in Higher Education. In: SIMPÓSIO BRASILEIRO DE INFORMÁTICA NA EDUCAÇÃO (SBIE), 37. , 2026, Goiânia/GO.
Anais [...].
Porto Alegre: Sociedade Brasileira de Computação,
2026
.
p. 802-814.
DOI: https://doi.org/10.5753/sbie.2026.27325.
