Hi, I'm an Applied Scientist and passionate about training multimodal LLMs on large-scale data. Currently working as an Artificial Intelligence Specialist at GAB and PhD Student at UTFPR in CPGEI, a program rated CAPES grade 7 (Brazil's highest), with a research extension at the University of Toronto. My research spans multimodal LLMs and computational social systems: how vision-language models perceive sentiment, and whether persona prompting produces valid variation in urban perception. I work with MInDS-Lab at UTFPR and the Critical Making Lab in Toronto, which hosts CIFAR-funded work on socially grounded AI safety.
More at neemiasbuceli.dev
|
Population Fidelity: Evaluating Population Representativeness in LLMs - 2026-09-28 Persona Prompting in Multimodal Urban Perception: Descriptive Convergence and Interpretive Variation - 2026-08-07 Stable Behavior, Limited Variation: Persona Validity in LLM Agents for Urban Sentiment Perception - 2026-04-30 Grandes Modelos de Linguagem Multimodais (MLLMs): Da Teoria à Prática - 2026-02-11 Multimodal LLMs See Sentiment - 2025-08-23 A convolutional neural network approach for counting and geolocating citrus-trees in UAV multispectral imagery - 2020-02-18 Regression in Convolutional Neural Networks applied to Plant Leaf Counting - 2019-09-09 More on Google Scholar |
CultureFinetuneSteerMLLM - 2026-09-27 Persona-Interpretive-Analysis - 2026-09-19 MLLMs-Teoria-e-Pratica - 2026-09-11 multimodal-LLMs-see-sentiment - 2026-08-23 MLLMs-persona-evaluation - 2026-08-23 More on GitHub |
