AI-Assisted Systematic Reviews of Medical Literature
We develop multi-agentic systems to support Evidence-Based Medicine.
Research group CIAAM-USP
Agentic
Complex
Systems
We study emergent phenomena and capabilities within and across generative systems trained for natural language communication.
01
Research directions
Our current projects
AI Safety and Alignment Auditing
We develop methods to evaluate robustness, harmful prompt detection, and alignment dynamics in large language models.
Collective Intelligence in Multi-Agent Systems
We analyze coordination, cooperation, and emergent behavior in interacting agents at scales beyond individual human oversight.
Fundamentals of Machine Learning
We study fundamental questions in machine learning, such as learning dynamics, limits on representational capacity, and domain adaptation methods.
02
Publications
Papers, preprints & conference contributions
2026
Automated Risk of Bias Assessment in Systematic Reviews Using Large Language Models: A Multi-Model Single-Agent Benchmark on 227 Randomised Controlled Trials
Kolmogorov-Arnold Networks for Small Language Models
Robustness of Language Models against Portuguese Harmful Prompts
Persona-Model Collapse in Emergent Misalignment
2025
Moral Susceptibility and Robustness under Persona Role-Play in Large Language Models
Deceive, Detect, and Disclose: Large Language Models Play Mini-Mafia
03
People
The researchers behind the work
Coordinators
Prof. Nestor CatichaCoordinator
Prof. Renato VicenteCoordinator
Postdocs
Dr. Felippe AlvesPostdoctoral Researcher
Dr. Luiz F. Guain TeixeiraPostdoctoral Researcher
PhD Students
Felipe RubimPhD Student
Estevão VieiraPhD Student
Master Students
Gabriel CostaMaster Student
Sergio BoniniMaster Student
Lucas PortaMaster Student
Gustavo MezzovillaMaster Student
Leonardo ValadaoMaster Student
Giovanni DalviMaster Student
Carlo LemosMaster Student
Undergrad Students
Thais MartinsUndergraduate Student
Thainara AssisUndergraduate Student
Gustavo BernardoUndergraduate Student
Gabriel MattosUndergraduate Student
Vinicius GuimarãesUndergraduate Student
Marina ReuterUndergraduate Student
Inácio CapituleioUndergraduate Student
Henrique TavaresUndergraduate Student
Matheus QueirozUndergraduate Student
Rafael CamõesUndergraduate Student
Lucas MaiaUndergraduate Student
Associated Researchers
Dr. Davi Bastos CostaUniversity of Chicago
Prof. Rodrigo VeigaUniversity of Nottingham
Dr. Felipe PoloUniversity of Michigan
Prof. Luiz MormilleUniversity of Soka
Eduardo AmorimUFPE
Rodrigo M. Cabral-CarvalhoUSP/UFABC
04
Open resources
Learn, discuss & explore
Teaching / September 2026
Neural Networks:
from Turing to LLMs
Open lecture by R. Vicente, IME-USP. Explore the ideas through interactive demonstrations.
Weekly discussions
Journal Club
Every week, we choose one paper—or a paper idea—as a starting point for discussion and idea generation.
Explore the discussions →05
Research insights
Ideas and perspectives from the group
June 2026
AI agent harnesses
June 2026
Learning at the Learning Conference: A Brief from ICLR 2026
May 2026
Emotion Concepts and their Function in a Large Language Model
April 2026
Large Language Models for Systematic Reviews: What Multi-Agent Approaches Can Teach Us
March 2026
Robustness of Language Models against Harmful Prompts in an Underrepresented Language
March 2026
Emergent Misalignment: sneaky general misbehavior from a few specific bad examples
06
Supported by
TELUS Digital Research Hub
Center for Artificial Intelligence
and Machine Learning (CIAAM-USP)