Machine learning researcher and professor at Insper. Audio, language, and AI for education.
Google Scholar | ORCID | Lattes | Blog | LinkedIn
I am an Assistant Professor at Insper (São Paulo). My work is in machine learning: representation learning for audio and text, and more recently, AI-driven tools for education.
My current research centers on three areas: multimodal representation learning (building models that jointly understand audio and language, including contrastive pre-training and zero-shot transfer); ML foundations (measuring structure and similarity in embedding spaces); and AI in education (LLM-driven formative feedback at scale, learning analytics, and data-driven equity analysis).
I am open to collaborate with researchers in signal processing, NLP, learning analytics, and data-driven decision-making.
PreprintTiago F. Tavares, Fabio Ayres, Paris Smaragdis, "Diagnosing Neural Convergence with Topological Alignment Spectra". . https://arxiv.org/abs/2411.08687
PreprintTiago Fernandes Tavares, "LogiPart: Local Large Language Models for Data Exploration at Scale with Logical Partitioning". . https://arxiv.org/abs/2509.22211
PreprintTiago Fernandes Tavares, Fábio José Ayres, "Multi-label Cross-lingual automatic music genre classification from lyrics with Sentence BERT". arXiv. 10.48550/ARXIV.2501.03769
ConferenceTiago Fernandes Tavares, Fabio José Ayres, Zhepei Wang, Paris Smaragdis, "On Class Separability Pitfalls In Audio-Text Contrastive Zero-Shot Learning". ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP). 10.1109/ICASSP49660.2025.10888245
ConferenceTavares, Tiago Fernandes, Soares, Luciano Pereira, "Automated Formative Feedback for Short-form Writing: An LLM-Driven Approach and Adoption Analysis". 2025 International Joint Conference on Neural Networks (IJCNN). 10.1109/IJCNN64981.2025.11229260
ConferenceTiago F. Tavares, Fábio Miranda, "Explorando o Gap de Gênero em Matemática no ENEM em Municípios Brasileiros como Indicativo de Potenciais Candidatas para Carreiras em Computação". Anais do XXXII Workshop sobre Educação em Computação (WEI 2024). 10.5753/wei.2024.3003
PreprintTiago Fernandes Tavares, Fabio Miranda, "Searching for winning study strategies for the Brazilian ENEM Exam". Center for Open Science. 10.35542/osf.io/mrwtd
ConferenceTiago Fernandes Tavares, Marcio Fernando Stabile Jr., "Enhancing Learning Effectiveness with Online Educational Platforms: Insights from the Use of Prairie Learn in a Data Science Classroom". PAAE/ALE.
ConferenceZhepei Wang, Cem Subakan, Krishna Subramani, Junkai Wu, Tiago Fernandes Tavares, Fabio Ayres, Paris Smaragdis, "Unsupervised Improvement of Audio-Text Cross-Modal Representations". 2023 IEEE Workshop on Applications of Signal Processing to Audio and Acoustics (WASPAA). 10.1109/WASPAA58266.2023.10248180
ConferenceTiago Fernandes Tavares, Lucas Bertoloto, "Air Drums, and Bass: Anticipating Musical Gestures in Accelerometer Signals with a Lightweight CNN". 2023 IEEE 33rd International Workshop on Machine Learning for Signal Processing (MLSP). 10.1109/MLSP55844.2023.10285920
JournalTiago Fernandes Tavares, José Eduardo Fornari, "Finite State Machines with Data Paths in Visual Languages for Music". Computer Music Journal. 10.1162/comj_a_00688
JournalGeise Santos, Tiago Fernandes Tavares, Anderson Rocha, "Reliability and generalization of gait biometrics using 3D inertial sensor data and 3D optical system trajectories". Scientific Reports. 10.1038/s41598-022-12452-6
JournalAugusto Amstalden, Tiago Fernandes Tavares, Anésio Azevedo Costa Neto, Giovana Cardi Camarini, "Audiovisual interactive artwork via web-deployed software: Motus composes Homino-idea". REVISTA VÓRTEX. 10.33871/23179937.2022.10.1.6
JournalTiago Fernandes Tavares, "Open-set classification approaches to automatic bird song identification: towards non-invasive wildlife monitoring in Brazilian fauna". IEEE Latin America Transactions. 10.1109/TLA.2022.9904764
JournalGeise Santos, Marcelo Wanderley, Tiago Fernandes Tavares, Anderson Rocha, "A multi-sensor human gait dataset captured through an optical system and inertial measurement units". Scientific Data. 10.1038/s41597-022-01638-2
ConferenceGeise Santos, Johnty Wang, Carolina Brum, Marcelo M. Wanderley, Tiago Fernandes Tavares, Anderson Rocha, "Comparative Latency Analysis of Optical and Inertial Motion Capture Systems for Gestural Analysis and Musical Performance". NIME 2021. 10.21428/92fbeb44.51b1c3a1
ConferenceMarcio Albano H Ferreira, Tiago Fernandes Tavares, "Continuous Parameter Control Using an On/Off Sensor in the Augmented Handheld Triangle". Computer Music Multidisciplinary Research.
ConferenceTiago Fernandes Tavares, Thales Roel Pessanha, Gustavo Nishihara, Guilherme Zanchetta L. Ávila, "Alloy Sounds: Non-Repeating Sound Textures with Probabilistic Cellular Automata". Digital Audio Effects - DAFx.
ConferenceThales Roel Pessanha, Thiago Rossi Roque, Guilherme Zanchetta L. Ávila, Lucas B Pereira, Gabrielly Lima De Oliveira, Bruna C. M. Pinheiro, Renata F. P. P. Paulino, Tiago Fernandes Tavares, "InFracta: the Body as an Augmented Instrument in a Collaborative, Multi-Modal Piece". International Conference on Computer Music.
Email: tiagoft1 at insper dot edu dot br