Guilherme Shinohara, a student at the Federal University of Mato Grosso do Sul (UFMS), presented the paper “Small Language Models for Real-World Task Automation in Resource-Constrained Environments” at ENIAC 2026. The work explores how small language models (SLMs) can support practical task automation when computing resources are limited.
The study evaluates the balance between response time and task performance when running SLMs on constrained hardware, and compares their outputs with those of larger language models. This line of research can help identify when smaller models are a useful option for applications that need to run locally or within limited computational budgets.
ENIAC is the Brazilian National Meeting on Artificial and Computational Intelligence. Its 23rd edition, ENIAC 2026, takes place in Cuiabá alongside BRACIS 2026, bringing together researchers and students to share work in artificial intelligence and computational intelligence.
The paper’s authors are H. R. B. Gonçalves, P. B. G. Neto, Guilherme Shinohara, J. de A. Silva, and Brivaldo A. da Silva Jr.