Preliminary program: sessions, times and rooms may still change.
Every session, paper and invited talk, by day, time and room.
Pantanal Auditorium
Chair: Daniela Barreiro Claro, UFBA
Balancing Coherence and Granularity in Legal Topic Modeling: A BERTopic-based Study on Portuguese Appellate Court Decisions
Fusão Híbrida para Recuperação de Informação Jurídica Brasileira
Sector-Specific Financial Sentiment Lexicons for Portuguese: Evidence from Brazilian Stock News
Do LLM Judges Agree with Humans? Evaluating Financial Commentaries from Material Facts
Cerrado Auditorium
Chair: Dennis Balreira, Universidade Federal do Rio Grande do Sul (UFRGS)
moBERTo: A Modern Encoder for Portuguese via Continued Pretraining of ModernBERT
BERTomelo: Your Portuguese Encoder Best Friend
Tokenizer Impact on Multilingual to Portuguese Transfer in ModernBERT-Based Models
TECOBEE: Adaptive Acceleration BERT Inference via Text Complexity-Based Early-Exit
Diffused Embeddings: Exploring Sentence Representations from DLMs
Language Organization in Multilingual LLMs through SAE Feature Suppression
Interpretability of LoRA Adapters: A Three-Way Dissociation
Faithfulness in Black-Box LLM-Based Recommender Systems’ Explanations
Nobres Room
Chair: Marco Aurélio Rocha, Universidade Estadual Paulista "Júlio de Mesquita Filho" (UNESP)
Learning to Order: GNN-Based Deep Reinforcement Learning Policies for Autoregressive Graph Models
DensGNN: A Density Based Graph Neural Network for Node Clustering in Applications with Noise
Benchmarking Spatio-Temporal Graph Neural Networks for Time Series Forecasting Across Heterogeneous Graph Domains
Multi-Domain Geometric Representation Learning for Structured Events
Crime Signature: Gang Graffiti for Criminal Risk Prediction in Street Networks
Contrastive Deep Graph Clustering with Dynamic Cluster Estimation
Strict core of a friend-appreciation hedonic game as a clustering solution
Hypergraph-Based Affinity Degree Detection for Researchers and Government Strategic Themes
Araguaia Room
Chair: Robson Rogério Dutra Pereira, IFMT Campus Cuiabá
Influencer Recommendation From Campaign Briefings: A Briefing-Centric Approach Using Sparse Representations And Density-Based Clustering
Rethinking The Evaluation Of Data Science And Data Management Research
Behavioral And Acoustic Artist Graphs For Music Community Detection And Genre Inference
Automated Information Extraction From Forensic Medical Reports Using Ocr And Rule-Based Methods
Zero-Shot Recognition Of Utility Meter Readings Using Vision-Language Models And General-Purpose Ocrs
Defi Protocol Risk Analysis: An Approach Based On On-Chain Metrics And Classifier Committees
Application Of Machine Learning Ensembles To Oil Production Prediction
Calibration Of Low-Cost Air Quality Sensors Using Machine Learning Models
Xingu Room
Chair: Anderson R. Correia Buarque da Silva, SENAI CIMATEC
A Data-Centric Framework for OOD-Robust RUL Prediction of Lithium-Ion Batteries with Risk-Aware Evaluation
A Selective and Regularized Extension of Gaussian Naive Bayes for Conditional Independence Relaxation
Prediction System for National Football Team Match Results
QL-Implications: Quantum-Fuzzy Modeling and Application to Financial Decision-Making
Feature Distribution and Co-occurrence Patterns in Public Credit Decision Datasets
Burnout Risk Prediction with Machine Learning and an Interactive Interface for Organizational Use
Unsupervised Learning for Identifying Clinical and Symptom-Based Profiles in Endometriosis: A Systematic Literature Review
Beyond Candidate Coverage: 3D Point-Profile Re-scoring for Individual Tree Detection in Agroforestry
Vila Bela Room
Chair: Jomi Fred Hübner, Universidade Federal De Santa Catarina
Between Polarization and Conversion: The difference between Herd and Framing Effects in Social Networks through Agent-Based Simulation
Decision Support Systems Based on Multi-Agent Systems for Tabletop Role-Playing Games: A Systematic Literature Review
Evacuation Modeling in Flood Scenarios Using Agent-Based Modeling: A Systematic Literature Review
From Cognition to Simulation: A Systematic Mapping of Visualisation and Interaction in Multi-Agent Systems
Applying Q-Learning to Herding Behavior: A Single-Agent Approach in NetLogo
Emergency Evacuation Analysis in a School Building via Agent-Based Simulations
Systematic Review of Reliability and Trustworthiness on Multi-Agent Architectures
An Automated Approach for Verifying Learned Policies of Q-Learning Agents in NetLogo
Chapada Room
No activities in this room at this time.
KeynoteSTIL
Aline Villavicencio, University of Exeter and University of Sheffield, United Kingdom
Pantanal Auditorium
Pantanal Auditorium
Closed-Loop Evaluation of Bird’s-Eye-View Maps from Cross-View Transformers as Inputs to Behavior-Cloning Policies
Comparing TF-IDF, Sentence-BERT, and LSTM for Self-Admitted Technical Debt Classification
A Low-Cost, Open Platform for End-to-End Autonomous Driving on a Miniature Ackermann Vehicle
Multi-Agent Planning with MCP Tools for Web Form Generation
Data-Driven Metric Learning for Image Databases Represented as Riemannian Manifolds Embedded into Image Space
Benchmark of XAI Methods for Image Classification and Extension of Optimization-Based Explanations to CLIP Models
Leveraging Task-Adaptive Pretraining for Misogynistic Hate Speech Classification in Brazilian Portuguese
Systematic Review of Quantum Machine Learning for Healthcare
Inferring Property Typology in Real Estate Tax Records via Domain-Aware Feature Alignment
XAI-Driven Loss Optimization via Activation Map Alignment in Multi-Task CNNs: A Systematic Literature Review
Surprise! How Do Science-Fiction Stories Preserve Narrative Tension Under AI-Assisted Rewriting?
Rethinking Solar Forecasting: Why Univariate Transformers Challenge Classical Statistical Models in Multi-Series Data
Large Language Models as Instrument-Aware Guardrails for Structured Mental Health Questionnaires
Evaluating Quantized Small Language Models in a TableRAG Pipeline for Table Question Answering
An Architecture for Integrating LLM-Based Agentic Systems with Multi-Agent Simulations
A Production-Oriented Machine Learning Pipeline for Real Estate Mass Appraisal with Official Geospatial Feature Engineering
Virtual Reporter: Automation of News Report and Journalistic Data Collection Using LLMs
Embedded Vision AI for UAV-Based Weed Detection in Sugarcane Field Using Semantic Segmentation
A Synthetic Ground-Truth Dataset for Online Map Matching
Evaluating Document Pre-processing Techniques for LLM-based Information Extraction
Diversity-Driven CNN–Transformer Ensemble for Robust Single-View Mammography Screening
A Equalização de Assinaturas Keystroke Oriundas de Teclados Mobile e Físicos Como Instrumento de Viabilização de Interoperabilidade
Evaluating Table Extraction Strategies for RAG Pipelines over Portuguese PDF Documents
Behavioral Finance Model: Google Trends and LSTM in Forecasting Retail Assets on the Brazilian Stock Exchange
MulSHEP: MultiStage Health Extraction Pipeline LLM for Brazilian EHR Data
Parameter-Efficient Coffee Leaf Disease Classification Using Quaternion-Valued EfficientNets
Active Adaptation for Novelty Detection in Data Streams with Extreme Latency Using Fuzzy Micro-Clusters
Less is More: Improved Fact-Checking on AVeriTeC with Iterative Question Generation
Meta-Learning-Based Algorithm Selection for Multi-Objective Ridesharing-Inspired Optimization
Animal Monitoring Systems in Real-World Environments: A Literature Review of IoT and Computer Vision Technologies
Disorder-Oriented Genetic Algorithm for Exploration Optimisation in Robot Swarms
An Optimizable Retrieval-Augmented Generation Workflow for Entity Extraction in Portuguese Fraud-Investigation Documents
Task-Adaptive Wavelet Decomposition with Quadrature Mirror Filters for Time Series Classification
Issue Generation from Consumer Complaints through LLMs
Evaluating Speech-to-Text Models for Brazilian Portuguese in Égua Assist: Supporting Programming Learning for People with Visual Impairments
Sentiment Analysis in Psychotherapy in Portuguese Idiom: The Role of Domain‑Specific Fine‑Tuning
Evaluation of Tiling and SAHI Strategies for Symptomatic Region Detection in UAV Images of Coffee Plantations Using YOLO
What Do Guests Really Think? A Multi-Scenario ABSA Framework in Brazilian Amazon Tourism
Multi-Head Attention and Dual Message-Passing Neural Network for Neural Algorithmic Reasoning
Hybrid Cognitive Systems in the Era of LLMs: Distributed Cognition, the Rise of "Vibe Coding'', and New Sociotechnical Risks in Software Engineering
Prospecting Potential Sites for Concentrating Solar Power (CSP) Plants through Embedding Similarity of the AlphaEarth Foundations Model
Analysis of a Generative Adversarial Network and a Denoising Diffusion Probabilistic Model for the Design of Nuclear Reactor Loading Patterns
An Experimental Comparison of Multi-Agent and Monolithic Prompt Architectures Across LLM Scales for Hardware Diagnostics
Application of Retrieval-Augmented Generation (RAG) in the Automated Extraction of Legal Arguments in Labor Administrative Defenses
Evaluating Imputation under Missingness Mechanisms for Credit Risk Modeling
Unsupervised Reconstruction of Abdominal Muscle Contours in CT Images based on Statistical Shape Modeling
Comparative Evaluation of Prompt Compression Techniques for Large Language Models: Extractive, Abstractive, and Token-Pruning
Deep Learning for Lung Cancer Histopathological Image Classification: Exploring Data Augmentation and Ensemble Learning Strategies
Efficient LoRA Adaptation of SAM3 for Interactive Bone Segmentation in Visible Human Cryosection Images
Evaluation of Projected Gradient Descent in Attacks against Age Estimation Models
An Experiment on the Simulation of Deterministic Turing Machines by Large Language Models
Computer Network Security Report Generation with RAG and Small Language Models
Generate or Rerank? Where to Invest an LLM’s Budget in Cold-Start E-Commerce Recommender Systems
Artificial Intelligence for Virality Analysis on Short-Form Video Platforms: A Taxonomy, Challenges, and Research Agenda
Comparing LSTM, GRU, and Transformers in Traffic Forecasting Across Varying Context Sizes and Horizons
Automated Audit of Judicial Metadata with Natural Language Processing: A Multi-Level Cascade Approach
Estimating Forecast Difficulty in N-BEATS using Latent Representations and Meta-Learning
Hyperdimensional Computing for Fake News Detection: A High-Efficiency Approach for Large-Scale Data
CNNs, Vision Transformers, and Ensembles for Glaucoma Classification: A Comparative Analysis
Analysis of the Sustained Performance of Deep Learning Models for Driver Assistance in Mobile SoCs
High-Sensitivity Bone Tumor Classification in Radiographs Using Deep Learning and Ensemble Learning Strategies
Predicting Asthma Exacerbations in Digital Health Ecosystems Using Machine Learning Frameworks
Structure-Preserving Extraction of Tabular Data from Engineering Manuals
Model-Based Reasoning under Bounded Parameter Uncertainty: Polynomial-Chaos Robustness of Normative Power-Quality Verdicts
The Impact of Data Imbalance on the Explainability of Text Classification Models: A Comparative Analysis with SVM and LSTM
TU-GUA: Architecture with Transformer and Unified Attention for Bone Tumor Segmentation in Radiographs
Guardrails in Large Language Models for Domain-Specific AI Assistants: A Systematic Literature Review
Integration of Niching Strategies into the C-DEEPSO Algorithm: An Approach for Multimodal Optimization
Improving Diabetic Retinopathy Lesion Detection and Segmentation via a Macula-Guided Region of Interest Pipeline Based on YOLOv12
Pattern identification in academic trajectories within the Brazilian Military School System using machine learning
A Deep Learning Approach for Soybean Grain Quality Analysis Using Multispectral Images
DISCALCULAI: AComputational Approach for the Identification of Dyscalculia in Educational Contexts
Driver Drowsiness Detection Using Computer Vision with Adaptive Methods on Android
Count Time Series Forecasting Applied to Crime Series in a Multi-Neighborhood Urban Region
Computer Vision for Automated FAMACHA Classification in Sheep and Goats: A Comparison of Color-Based and Deep Learning Strategies
Vegetation Dynamics Forecasting Using Multispectral Time Series: A Comparative Study of Statistical, Machine Learning, and Deep Learning Models
Open-Weight Language Models for Long-Context Academic Summarization
Machine Learning for Glaucoma Identification: A Comparative Study of Algorithms
4SkeletonFeat: A Framework for unifying the extraction of features from videos of human activities
Cerrado Auditorium
Bruno Padilha, USP
Cerrado Auditorium
Marlesson Santana, Jusbrasil
Cerrado Auditorium
Nobres Room
Chair: Paulo Ferreira, Universidade Federal de Pelotas (UFPEL)
From Bias to Fairness: Can Synthetic Data Bridge the Gap in Tabular Machine Learning?
FairMOP: Benchmarking Fairness-Utility Trade-offs in Text-to-Image Models via Pareto Frontiers
Coverage Before Recognition: Auditing Architectural Erasure in Face Detection
Open-set Zero-Day Release authorship attribution of code for Digital Forensics
Improving Text Classification via XAI-Guided Misclassification Augmentation
Bisection-Based Neighborhood Generation for Local Rule-Based Explainable Artificial Intelligence
BAH.IA: Translating National AI Guidelines for Basic Education into a Scalable Framework for Contextualized AI Learning
Time to Rethink NER Splits
Araguaia Room
Chair: Elloá B. Guedes, Universidade do Estado do Amazonas (UEA)
Machine Learning-Based Clinical Triage For Leptospirosis Using Sinan/Sus Data
Binary Screening Gate Normal/Abnormal In Brazilian Upper Gastrointestinal Endoscopy: Safety Analysis Guided By Target Sensitivity
Clinically Interpretable Risk Stratification In Chagas Disease Using Novel Derived Variables And Calibrated Decision Trees
Knowledge Distillation And Composite Loss As A Low-Resource Strategy For Ldct Image Denoising
Evaluating Tabpfn V3 For Multi-Label Toxicity Prediction: A Comparative Study On Tox21 And Sider Datasets
Evaluating Prompting Strategies For Clinical Information Extraction In Portuguese Using Langextract
Biochemical Knowledge Extraction Through Similarity Search Via Finetuned Open-Source Large Language Models
Multiscale Forecasting Of Dengue Alerts From Notifications, Historical Seasonality, And Socio-Spatial Vulnerability
Xingu Room
Chair: Adriana Pagano, Universidade Federal de Minas Gerais (UFMG)
Curupira: Arquiteturas Híbridas Transformer-BiLSTM para Detecção de Discurso Ofensivo em Português Brasileiro
MultiDomainYT-ToxBR: Um Corpus Multidomínio de Comentários Tóxicos do YouTube em Português Brasileiro
Evaluating LLM-based Triple Extraction for Knowledge Graph Fact-Checking in Portuguese
Autonomous Fact-Checking: Integrating Local Inference and Adversarial NLP Attacks
Tail Smoothing and Cross-Lingual Volatility: Evaluating Estimative Uncertainty in Large Language Models for Brazilian Portuguese
A Comparison of Commonly Used Automatic Evaluation Metrics for Open-Ended Tasks in Portuguese
Vila Bela Room
Chair: Eder Mateus Nunes Gonçalves, Universidade Federal do Rio Grande (FURG)
Unifying Graph-Based Retrieval: A Higher-Order Topological Framework for RAG
RegMem: Graph-Based Regulatory Memory for Faithful Multi-Turn RAG in Safety-Critical Domains
A Comparative Analysis of RAG Architectures for Diagnostic Support Using Telemedicine Data
Noise-Gated Retrieval: Embedding-Based Context Compression for Cost-Aware RAG
Prompt-Only LLM Routing with Conformal Regret Control
Grammatical Representations in Transformer Attention: A Multidomain Study of Portuguese via UD
Fixed-Content Perplexity as a Low-Overhead Contextual Boundedness Score for LLM Systems
Beyond Multilingual Averages: MTEB-PT, a Benchmark for Portuguese Sentence Encoders
Chapada Room
Prof. Dr. Julio C. dos Reis, Institute of Computing, University of Campinas, Brazil
André Gomes Regino, CTI Renato Archer
Chapada Room
Pantanal Auditorium
Chair: Heraldo Pimenta Borges Filho, IRISA
Knowledge Graph-Guided Entity Linking With Llms For The Identification Of Construction Cost Items
Distilling Llm Routers Into Compact Classifiers Under Data Scarcity In Intelligent Tutoring Systems
Customer Experiential Knowledge Classification In Online Tourist Reviews With A Large Language Model
Class Imbalance And Batch Effects In Llm-Based Screening For Systematic Reviews
Manuscribe Ai: Learning Support System For People With Fine Motor Impairments
Mapping Researchers To Strategic Themes With Language Models Via Reasoning- And Reward-Guided Fine-Tuning
Retrieval-Augmented Generation For Question Answering Over Brazilian Portuguese Institutional Normative Documents
Multimodal Pictogram Embeddings For Augmentative And Alternative Communication: A Comparative User Study
Cerrado Auditorium
Chair: Claudio Elizio Calazans Campelo, Universidade Federal de Campina Grande (UFCG)
PULSAR: A Multi-Agent System for Monitoring Hospitalized Patients
Balancing Immediate Responsiveness and Strategic Planning in Primary Health Care: a human-in-the-loop solution to multi-horizon demand forecasting
Classifying Psychological Distress in Brazilian Adolescents: A Multi-regional Machine Learning Approach
Can low-parameter LLMs provide trustworthy simulations of human BDI-II responses? A Comparison of Biodemographic and Instagram-Based Context Profiles
Ethos AI: Artificial Intelligence Agents Based on LLM as Simulated Patients for Training Psychologists in the Treatment of Alcoholism
Segmentation and Classification in Therapeutic Transcripts with Fine-Tuned BERT and SBERT
MarIA: A Multimodal Approach to Voice Accessibility in Primary Health Care
VygotEA: A Hybrid AI System for Adaptive Socioemotional Support in Students with ASD
Nobres Room
Chair: Fábio Gagliardi Cozman, Universidade de São Paulo (USP)
A Neuro-Symbolic Framework for Logic Puzzle Solving: Pseudo-Boolean Reasoning with LLM-Based Explanations
Bottom-Up Knowledge Compilation with And-Sum Circuits: Theoretical and Empirical Results
A Neuro-symbolic reranking pipeline for level-aware learning-object recommendation: A case study in School-Principal Training
Supervised and Zero-Shot Creative Thinking Assessment in PISA 2022 Portuguese Texts
A Multi-Level Interpretable Framework for Skin Cancer Classification Using Deep Learning and Fuzzy Systems
Continuous Ranking from Binary Relations: An Unsupervised Structure Learning Approach to Passport Mobility
Research–Theme Affinity Prediction via Early Embedding Fusion Operators
Rare Signal Distillation in Comments on Sponsored Videos: Brand Ubiquity and Thematic Engagement on YouTube
Araguaia Room
Chair: Kauan Divino Pouso Mariano, Universidade Federal de Goiás (UFG)
Context-Aware Recommendation System for Augmentative and Alternative Communication
Comparison of Deep Learning Models in the Detection of Fraud in Scientific Images
A Hybrid Approach for Brazilian Coffee Scene Classification Using VGG16 and Image Features
Task2Vec Readiness: Diagnostics for Federated Learning from Pre-Training Embeddings
Benchmarking Local Large Language Models for an Intelligent Tutoring System in Digital Logic
Audio and Voice Deepfake Detection: a Survey on Cross-Dataset Generalization Bottlenecks
Prototypical Few-Shot Classification with Vision-Language Models: A Taxonomy and Systematic Study
Xingu Room
Chair: Jackson Wilke da Cruz Souza, UFBA
Vidya Syl: A syllable division algorithm for Brazilian Portuguese adhering to the Common Orthographic Vocabulary of the Portuguese Language
CorPop26: Uma atualização das 3.000 Palavras Mais Frequentes do CorPop de 2018
Extending Enhanced Universal Dependencies to Nheengatu: Semi-Automated Control and Raising Annotation
Clonagem de Voz por IA para o Português Brasileiro em Contexto Assistivo: Desafios Comparativos em Cinco Ferramentas de Síntese de Fala
Leakage-Aware Evaluation of Brazilian Clinical Notes for ICU Mortality Prediction: A Study on the BRATECA Dataset
Exploring Hybrid Pre-training for Automatic Essay Scoring
Vila Bela Room
Chair: Julio Cesar dos Reis, UNICAMP
MATE: An Automated Organization-Driven Testing Environment for JaCaMo Multi-Agent Systems
Modelling Institutional Enactment in Multi-Agent Systems with Enact-as Rules
Multi-Agent Systems and Prompt Engineering in automating and optimizing prompt generation and refinement processes: a Systematic Literature Review
Stagent: A Finite-State Machine Architecture for Traceable LLM Agents
Jaime Simão Sichman, USP
Vila Bela Room
Chapada Room
Eduardo Laurentino, Itaú
Chapada Room
KeynoteBRACIS
Marcelo Coelho, MIT Design Intelligence Lab
Pantanal Auditorium
No activities for the selected events on this day.
Pantanal Auditorium
Synapse: Automated Generation of Systematic Review Protocols and AI-Assisted Screening with Large Language Models
Oral Lesion Classification Using Deep Learning with ConvNeXt-Base
Exploratory Landscape Analysis as Meta-Features for Hyperparameter Tuning Recommendation: A Study on SVMs
Cesarean Delivery Prediction in Brazil Using SINASC 2022 Microdata: A Data-Centric Analysis of Social, Healthcare, and Obstetric Factors
Surrogate Models for Pipeline Recommendation in Multi-Label Classification
Comparative Evaluation of Feature Extraction Methods for GWAS-Based Alzheimer’s Disease Classification
How Well Do Small Language Models Understand the Brazilian Educational Context? An Evaluation on ENEM, FUVEST and Unicamp
Chunking Strategies for Retrieval in Enterprise Documents: A Comparative Study and the QI-RAG Proposal
Small Language Models for Real-World Task Automation in Resource-Constrained Environments
A Multidimensional LLM Evaluation Pipeline for Automatic Responses to Restaurant Reviews in Brazilian Portuguese
An Image-Based Urban Flood Detection System using Deep Learning and Conformal Prediction for Edge Deployment
Handling Noisy Labels in Industrial Scenarios: A Data-Centric Approach for Lubrication Fault Diagnosis
Arandu: A Multiagent Intelligent Tutoring System Architecture with Generative AI for Gamified Teaching of Heuristic Evaluation in HCI
Instance Segmentation via Deep Learning for Machinery Monitoring in Industry 5.0
Sensitivity Analysis of Retrieval and Routing Parameters in Educational RAG Systems
Benchmarking Foundation Models for Zero-Shot Water Demand Forecasting
Evaluating the Impact of Textual Attributes on Distance-Based Learning from Heterogeneous Data Using Text Embeddings and PCA
Beyond Vector Search: Comparing Classical RAG with Hybrid GraphRAG for Climate Science Q&A
A Comparison of Incremental Semantic Segmentation Algorithms for the Recognition of Agricultural Threats from UAV Images
A Study of Continuous Regression for Diabetic Retinopathy Grading with Intra- and Cross-Dataset Evaluation
Reproduction Operator Sensitivity in MOEA/D under Tchebycheff Decomposition: An Empirical Study on Regular and Irregular Pareto Fronts
SeqLog: Surface-Invariant Detection and Event Localization in System Logs
Architecture over Scale: Multi-Agent Reasoning for ENEM Mathematics
A Normative Variational Autoencoder Scores Per-Cell Vulnerability in Parkinson’s Disease
A Client Selection Approach Based on Dynamic Clustering and Reputation in Federated Learning
Vibration-based Structural Health Monitoring for beams under temperature variation using integrated acceleration, velocity and displacement features
CRSLLM: A Hybrid Conversational Recommender System Integrating Collaborative Filtering and Large Language Models
A Comparative Analysis of Machine Learning Models for Acoustic Classification: A Reproducible Modular Pipeline Across Three Domains
Generation of Synthetic Data Based on Real Images for the Detection of Plastic Waste in an Urban Canal
VigiaEnchente: Static Flood Susceptibility Scoring, Validity Analysis and Risk-Penalized Routing in Urban Road Networks
Cross-Media Recommendation: A Multimodal Approach in Vector Recovery Systems
Detection of Planting Failures in Eucalyptus Stands Using Spectral Indices and Machine Learning on Sentinel-2 Imagery
Explainable AI for early detection of Antimicrobial Resistance: An Approach to Reduce Clinical Risk
A Transparent Multiclass Classification Architecture for Credit Variable Categorization
No-Transaction Band Action Parameterization in Deep Reinforcement Learning for Option Hedging under Transaction Costs
Sales Forecasting Using Time Series Models with Exogenous Logistics Variables
Evaluating Racial Fairness in Credit Granting Models
The Impact of Transformed Data Representations on Sensor-Based Human Activity Recognition
Mitigating Imbalance in Portuguese Toxicity Classification via LLM Augmentation and Cross-Dataset Label Propagation
Robustness and Explainability of Machine Learning Algorithms for Autism Spectrum Disorder Screening
Comparative Evaluation of Large Language Models in the Text-to-SQL Task in a Real Domain-Specific Database
Random Survival Forest for Neonatal Survival Prediction in Preterm Infants from a Brazilian National Cohort of 2014 to 2022
Scaling Limits of the Dropout Inversion Attack: An Empirical Analysis Under Increasing Input Dimensionality
Critic-Based Self-Correction in Local Retrieval- Augmented Generation: An Empirical Evaluation in an Institutional Academic Chatbot
LLM optimization in embedded hardware: a systematic review
Evaluation of Denoising Architectures on a Reconstructed Ultra-Low-Dose CT Dataset
Análise Exploratória da Relação entre Consumo de Vídeos Curtos e Capacidade Atencional com Técnicas de Aprendizado de Máquina
Embedded Reinforcement Learning: Paving the Roads Towards Fuel Efficiency
Silent and Vocalized Speech Classification Using Surface Electromyography and Machine Learning
A Supervised Machine Learning Framework for Sports Championship Prediction: Case Studies in the NBA and FIFA World Cup
Enhanced Multi-Robot System Architecture with Distributed Planning
Dendrobot: Phytosanitary Diagnosis of Rubber Tree Tapping Panels via Semantic Segmentation with Deep Learning
Automating the Software Development Lifecycle with LLM Coding Agents: A Webhook-Based Approach
Assistive Object Recognition on Android with TensorFlow Lite
Impact of Hybrid Routing on Curriculum-Aware Multi-Collection RAG Architectures
Generative AI Dependency and its Effects on Introductory Programming Learning: An Experimental Study
A Comparative Analysis of LLMs in a FAQ Chatbot: A Case Study
A machine learning-based data matching pipeline for linking electronic invoices to public procurement processes
A Graph-Based Approach for Detecting Intermediary Accounts and Prioritizing Investigations in Money Laundering
Analyzing Student Perceptions in Course Evaluation: A Hybrid Approach Using Topic Modeling and LLM-Based Automatic Labeling
Mitigating Algorithmic Gender Bias in Brazilian Tech Salary Prediction
Estimation of leaf dimensions through image segmentation using a new annotated dataset
Soft Voting Vision Transformer Ensemble for Acute Myeloid Leukemia Cellular Classification
Benchmarking Ranking Frameworks for Algorithmic Fairness
EchoSystem: An Integrated Platform for Bioacoustic and Micrometeorological Monitoring
Galaxy Morphology Classification with Deep Neural Networks
Explainability of Predictive Models in Healthcare: An Analysis of SHAP’s Local and Global Explanation Plots
Impact of Segmentation Techniques on the Performance of Automatic Classification of Dermatological Lesions
CranioSynth: A Pipeline for Generating Synthetic Cranial Defects Through LLM-Assisted Parameterization and Anatomical Constraints
A Comparative Analysis of Predictive Models for Hospital Average Length of Stay
Pipeline Interpretável para Detecção de Doenças Foliares em Tomateiro com Inferência em Borda via Rust/ONNX
Lightweight CNNs for Edge Vision: A Cross-Benchmark Study of Robustness and Efficiency
Sampling Is Not Selecting: Verifiability as the Bottleneck of Inference-Time Scaling
Applying Answer Set Programming and Metaheuristics to the Volunteer Assignment Problem with Preferences
Predicting PRIO3 Stock Prices: PatchTST vs. LSTM
Bridging the Gap in Facial Expression Recognition: Real-Time Use and Human-in-the-Loop Data Collection
Application of Reinforcement Learning in the Development of Intelligent Agents for the Game of Othello
Analysis of CNN architectures for object detection in drone images with a focus on Aedes aegypti.
Embedded Vehicle Detection with YOLOv5n on ESP32-S3 for Urban Traffic Applications
A Deep Learning Framework for Urban Ecoacoustic Monitoring and Management: A Synthetic-to-Real CNN-Transformer Approach
Cerrado Auditorium
Chair: Claudio Elizio Calazans Campelo, Universidade Federal de Campina Grande (UFCG)
An Axiological Matrix For Brazilian Portuguese Bluesky Discourse On Ai In Education
Statistical And Machine Learning Methods For Auto Insurance Claims Prediction: A Survey
From Importance To Action: Evaluating Counterfactual Explanations Under Constraints In Brazilian Public Schools
Farmacia: A Rag-Based Pharmaceutical Assistant For Brazilian Portuguese Medicine Leaflets
Evaluating Joint Named Entity Recognition and Relation Extraction on a Novel Portuguese News Dataset
Jacv: A Benchmark For Legal-Argumentative Coherence Verification In Brazilian Portuguese
Assessing The Impact Of Ai Trism (Trust, Risk, And Security Management) Modelops On The Operational Behavior Of Generative Ai Systems
Hylex-Rag: Bridging The Lexical-Semantic Gap In Regulatory Compliance Retrieval
Nobres Room
Chair: Reinaldo A. C. Bianchi, FEI
Wildfire Detection in Satellite Imagery via Weakly Supervised Contrastive Learning
Cross-Biome Evaluation of Deep Learning Architectures for Semantic Segmentation of Deforestation in the Amazon and Caatinga
HABShield: Disentangling Scale, Domain, and Channel Interface in Foundation Model Transfer for Ocean-Color HAB Detection
Physics-Informed Machine Learning for Precipitation Nowcasting: Towards Efficient and Reliable Predictions
Comparing Open Set Recognition Approaches for Monitoring Agricultural Crops
Empowering Farmers through Privacy-Preserving AI: A Federated Learning Approach for Soybean Disease Classification
Edge AI or Offloading? A Multi-Dimensional Evaluation of Object Detection Deployment
Beyond Architectural Novelty: Robust Real-World Road Infrastructure Analysis
Araguaia Room
Chair: Roseli Francelin Romero, Universidade de São Paulo (USP)
Deep Reinforcement Learning For Battery Arbitrage With Price And Renewable Data
Compliance Verification In Administrative Sanctioning Procedure Flows: A Semantic Frame-Based Approach
Evaluating Instance Selection For Active Learning Using Label-Agnostic Instance Hardness Measures
Error Pca: A Density-Overlap Approach To Supervised Feature Extraction For Classification
Particle Competition For Pseudo-Labeling In Positive And Unlabeled Learning
The Importance Of Ablation Studies For Complex Nonparametric Causal Models
Weightless Neural Trees
Exact Machine Unlearning On Weightless Neural Networks
Xingu Room
Chair: Daniela Barreiro Claro, UFBA; Emerson Cabrera Paraiso, PUC-PR; Vládia Célia M. Pinheiro, UNIFOR
Bootstrapping Text Anomaly Detection with LLM-Generated Weak Supervision
Compression-Based Linguistic Complexity Metrics in Automatic Essay Scoring
Como Nossos Pais: Princípios Clássicos para a Avaliação de Sistemas de IA (Principles on Evaluation Methodology)
Prompt Engineering for Small Language Models: Evaluating ICL for Portuguese Sentiment Analysis
The Impact of Language and Portuguese Intralinguistic Variation on NL2SQL
Vila Bela Room
Chair: Sávio Teles, Universidade Federal de Goiás (UFG)
An approach with Visual and Tabular Mamba to multimodal medical data using Mixed Fusion
Vessel-UNet: Attention-Enhanced U-Net for Retinal Vessel Segmentation
Physio-PPT: Physiology-Anchored Patch-Order Self-Supervised Learning for ECG Classification under Low-Label Regimes
CardioKGImpute: A Knowledge-Graph-Guided Imputation of Multivariate ICU Time Series
Beyond Intensity: Visibility Graph Morphologies for Disease Detection in FTIR Spectra
Improving Low-confidence AlphaFold Protein Predictions Using Deterministic Local Optimization
A Novel Machine Learning Approach for Central Nervous System Tumor Classification from DNA Methylation
Learning from Negative Edge Quality in Graph-Based Drug–Target Interaction Prediction
Chapada Room
Chair: to be announced
Intelligent Techniques for Resource Allocation using Multi-Agent Systems and Multi-Criteria Methods: A Systematic Literature Review
Towards MASPY Interoperability with Heterogeneous Multi-Agent Platforms
Wumpus Verse: A Web Platform Prototype for Development and Evaluation of Agents in the Wumpus World
MagicArtifact: A Language-Agnostic Approach for External Artifact Development in the Agents and Artifacts Model
Towards a Visual Programming Framework for BDI Agents
Mitigating LLM Hallucinations in Code Generation for BDI Agents
Knowledge Graphs for Decoupling Intelligent Agents in WoT Industrial Systems
A Multi-Agent System for Detection and Prioritization of Research Gaps in Scientific Literature
KeynoteBRACIS
Marcos V. Treviso, University of Lisbon
Pantanal Auditorium
Pantanal Auditorium
Chair: to be announced
Prosa: Rubric-Based Evaluation of LLMs on Real User Chats in Brazilian Portuguese
LabelRoute: Routing by Instance-Label Hardness for Multi-Label Classification
Latent Fact-Checking: Detecting Misinformation through Activation Engineering
Density as a Scaling Principle: Adaptive Spatial Context for POI Embeddings
Characterizing Grokking via Topological Data Analysis in a Small Transformer Model
The Free Inference Dimension: A Complexity Measure for Zero-Collision Navigation under Hypothesis Mixtures
Cerrado Auditorium
Eduardo R. Hruschka, Itaú
Cerrado Auditorium
Nobres Room
Chair: Lucas Nunes Alegre, Universidade Federal do Rio Grande do Sul (UFRGS)
Towards Real-Time Nonholonomic Local Trajectory Planning Based on RRT using Many-core GPUs
Offline Policy Evaluation as a decision-support tool for designing Adaptive Experiments
Reinforcement Learning with Value of Perfect Information Exploration
Quantitative Event-Based Explanations for Policies in Stochastic Planning
Value Iteration for Risk-Sensitive Markov Decision Processes with h-Local Exponential Average Reward Criterion
MEROS-EVM: An open-set human activity recognition model using wearable sensors data
Memory-Aware Multi-Agent Architecture with LLM-based Orchestration for Conversational Recommendation
A Scalable Hierarchical Alert Framework for Driver Behavior Risk Detection Using Unsupervised Learning on Large-Scale Telematics Data
Araguaia Room
Chair: Roseli Francelin Romero, Universidade de São Paulo (USP)
Beyond Algorithmic Disagreement: Stabilizing Counterfactual Explanations In Tree-Based Models. A Case Study
Auditable Inference In Physics-Regulated Classifiers: Physicsonly/Neuralonly/Hybrid Modes And A Fuzzy Reading Of The Differentiable Physics Layer
An Empirical Study Using Lightweight Convolutions To Achieve Green Ai
From Abductive Explanations To Global Logical Rules For Node Classification In Sgcs
A Drift-Aware And Fairness-Aware Explainable Ai Framework For Major Pathogen Transitions
Comparing Explainability Frameworks For Accident Severity Prediction
An Xai-Based Auditing Framework For Detecting Shortcut Learning In Visual Cardiac Murmur Screening
Evaluating Subgroup Fairness In Deep Learning Models For Breast Cancer Tumor Segmentation
Xingu Room
Chair: Roney Lira de Sales Santos, Universidade Federal da Bahia (UFBA)
DANTE – A Treebank of Tweets in Brazilian Portuguese
Novo CorPop Saúde: Data Collection for Simplified Health Communication in a Brazilian Patient Information Leaflet Corpus
Beyond Aggregate Scores: A Diagnostic Analysis of Portuguese LLM Evaluation Suites
ENEM Essay Feedback Using LLM-Augmented Prompts
Developing Multilingual Multiword Expressions (MWEs) Resources: An Annotated Corpus of Fables following the PARSEME guidelines
Expressões multipalavras no discurso de fóruns sobre saúde mental: desenvolvimento de um corpus em inglês e português brasileiro
EpiCorpus-BR: A Named Entity Recognition Corpus for Epidemiological Documents in Portuguese
Vila Bela Room
Chair: Hilario Tomaz Alves de Oliveira, IFES
An annotated corpus and computational models for offensive language classification in tweets about the 2022 presidential elections.
Hybrid Retrieval RAG System for Clinical Decision Support in Portuguese
Hybrid Legal Precedent Retrieval via LLM-Based Document Expansion and Dense-Sparse Fusion
United Experts: Efficient Multi-Task Sentiment Analysis in Portuguese via Model Merging
LLM-Based Summarization of Academic Communities Using Relational Group Characterization
The Impact of Automated Reinforcement Learning Fine-Tuning on Advanced Prompt Engineering for Code-Based Reasoning
Representation Engineering in the Legal Domain: A Comparative Study of Linear and Non-Linear Probing for Initial Petition Classification
Reading Brazil: A Local Cost-versus-Accuracy Benchmark of Open OCR Engines on Brazilian Documents
Chapada Room
Cleber Gomes da Silva, Arquiteto de Soluções Senior da AWS para Governo; ex-aluno de Ciência da Computação da UFMT
Vinicius Soares Batista, Arquiteto de Soluções Senior da AWS para Governo
Chapada Room
KeynoteKDMiLe
Washington Luiz Miranda da Cunha, UNICAMP
Pantanal Auditorium
Pantanal Auditorium
Chair: Paula Christina Figueira Cardoso, Universidade Federal do Para (UFPA); Lilian Teixeira de Sousa, Universidade Federal da Bahia (UFBA)
A Comparative Study of Discourse Signal Proposals in Rhetorical Structure Theory
Anotação de Metáforas em Saúde Mental: Análise de Postagens Baseada no Procedimento MIPVU
Análise de comportamento do modelo NotebookLM em transcrição automática
Classificação de Textos com Matrizes de Similaridade: Um Estudo de Caso em um Corpus com Nomes de Produtos
Concatenative Synthesis for Brazilian Sign Language
Da Definição das Entidades à Anotação do Corpus: Desafios da tarefa no Direito do Consumidor
Data-Faithful Natural Language Generation for Explaining Fitness-Market Anomalies
Desafios Tipológicos na Construção de Treebanks: O Mapeamento Morfossintático da Língua Kubeo no Padrão UD
Esquecida no Churrasco: Um método para validação e correção responsável de traduções automáticas do BBQ para Português do Brasil
Expansão da Ferramenta Doccano para a Anotação de Análise de Sentimento Estruturada
Explorando a Integração de Múltiplas Modalidades Visuais no Reconhecimento Contínuo de Libras
Heurísticas vs. LLMs na triagem de tweets: avaliando pistas linguísticas para tarefas de PLN em português
Identificação e classificação de trechos racistas e sexistas em processos judiciais
Integração de Conhecimento Linguístico Estruturado em LLMs para a Tarefa de Sumarização Extrativa
OABench-Consumidor: Um benchmark para avaliação de modelos de linguagem em Direito do Consumidor
Unsupervised Subword Segmentation for POS Tagging in Low-Resource Agglutinative Languages
Uso de modelos de linguagem na identificação de relações de coerência da Rhetorical Structure Theory em textos de User-Generated Content
Cerrado Auditorium
Chair: Lucas S. Kupssinskü, Pontifícia Universidade Católica do Rio Grande do Sul (PUCRS)
Constructing an Optimal Behavior Basis for the Option Keyboard
Characterizing Human Semantic Navigation in Concept Production as Trajectories in Embedding Space
Boosted GFlowNets: Improving Exploration via Sequential Learning
Cooperative Sheaf Neural Networks
An Algebraic Structuring of Epistemic States for BDI Agents
A BDI-Based Multi-Robot Architecture for Continuous Planning and Execution
Nobres Room
Tatiane Nogueira Rios, UFBA
Anna Helena Reali Costa, USP
Denis Deratani Mauá, IME-USP
Diego Mesquita, FGV
Jaime Simão Sichman, USP
Leliane Nunes de Barros, IME-USP
Ricardo M. Marcacini, USP
Solange Rezende, ICMC-USP
Nobres Room
Araguaia Room
Chair: Anderson R. Correia Buarque da Silva, SENAI CIMATEC
Forecasting Notarial Stamp Revenue Using Machine Learning: A Comparative Study At The Court Of Justice Of Goias
A Multivariate Hybrid Arima-Ga-Svr Approach For Short-Term Solar Radiation Forecasting
Time Series Forecasting Of Public Revenues And Expenditures: A Comparative Analysis Of Individual, Local, And Global Approaches
Hierarchical Forecasting Of Spare Parts Components On A Supply Chain Network
Comparative Evaluation Of Daily Rainfall Forecasting Models Using Inmet Data
Spatio-Temporal Forecasting Of Deforestation In The Legal Amazon Using Lstm Networks Under An Isolated Regional Scale Approach
Xingu Room
Josiel Maimone de Figueiredo, UFMT
Xingu Room
Jackson Wilke da Cruz Souza, UFBA
Daniela Barreiro Claro, UFBA
Vládia Célia M. Pinheiro, UNIFOR
Raquel Meister Ko Freitag, Universidade Federal de Sergipe (UFS)
Emerson Cabrera Paraiso, PUC-PR
Xingu Room
Vila Bela Room
Chair: Carolina Gil Marcelino, Universidade Federal do Rio de Janeiro (UFRJ)
Random-Key Optimizers for Quadratic Unconstrained Binary Optimization
Electric Vehicle Routing as a Multi-Objective Evolutionary Problem
Performance Analysis of MOEA/D with Surrogate-Based Algorithms in Multi-Objective Problems
Robustness Analysis of Deep Reinforcement Learning Agents Under Visual Perturbations
M³-CIL: A Multidimensional Evaluation Protocol for Class-Incremental Learning
Evaluating Computational and Carbon Costs of Machine Learning: The Impact of Feature Dimensionality During Model Development
Chapada Room
Cleber Gomes da Silva, Arquiteto de Soluções Senior da AWS para Governo; ex-aluno de Ciência da Computação da UFMT
Vinicius Soares Batista, Arquiteto de Soluções Senior da AWS para Governo
Chapada Room
No activities for the selected events on this day.
Pantanal Auditorium
Chair: Anne Magaly de Paula Canuto, UFRN; Denis Deratani Mauá, USP
An Integer Linear Programming and Biased Random-Key Genetic Algorithm for the Sequence-to-Graph Alignment Problem with Vertex Label Modification
Understanding Catastrophic Forgetting: A Three-Axis Structural Analysis in Class-Incremental Learning
Model and Evaluation of an Intelligent Tutoring System with a Reinforcement Learning Agent
Alignment-Free Viral Metagenomics: External Validation of K-mer Based Ensemble Classifiers on an Independent Amazonian Dataset
Computational Models for Identifying Aedes aegypti Breeding Sites in Aerial Images: A Systematic Review
Unsupervised Ensemble Clustering for Fake News Detection in Brazilian Portuguese using Contextual and Traditional Embeddings
Automated Large-Scale Labeling of Brazilian Urban Traffic Images Using a Deep Learning Ensemble
MCC: Multi-Centroid Curriculum for Pre-defined Centroid-based Classifiers
Cerrado Auditorium
Chair: Murillo Guimarães Carneiro, Universidade Federal de Uberlândia (UFU)
Analyzing Language Models for Hate Speech Detection in Brazilian Political Content
Do LLMs Take Sides? Measuring Opinion Bias and Sycophancy via Multi-Turn Probing
On the Feasibility of Distilled Small Language Models for Agentic Text-to-SQL in Portuguese under Real-World Conditions
Beyond Constrained Decoding: Grammar-Aware Fine-Tuning for Small Language Models
Supervised Fine-Tuning of Large Language Models for Strategic Decision-Making in Poker
Decomposing Memorization Reduction in Privacy-Preserving Fine-Tuning of SLMs for CSIRTs
A Framework for Assessing Hard-Constraint Compliance in LLMs
Distilling Reasoning from Language Models for Ranking Evaluation
Nobres Room
Chair: to be announced
Graph-based Airspace Control and Path Optimization for Last-Mile Delivery Drones
On the Trail of a Green and Golden Generative AI: Assessing LLM Text Generation in Brazilian Portuguese
Leveraging LLMs for Explainable Recommender Systems: Exploring User Perceptions and Faithfulness in Generated Explanations
Araguaia Room
Chair: Marley Vellasco, Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio)
A Diversity-Driven Ensemble Based on Data Distribution and Parameter Heterogeneity for Wind Speed Forecasting
Beyond Complexity Invariance: An Adaptive Multi-Scale Distance for Time Series Classification
How effective are Generative Adversarial Networks for Time Series Representation Learning?
Multi-Model Comparison for Weather Nowcasting: Integrating Predictive Performance, Green AI and Explainability
StreamGen: An Interactive Web-Based Generator of Non-Stationary Data Streams
MetaDD: Early Concept Drift Detection via Meta-Learning
F4RM: Feature-Restricted Round-Robin Random Mapping for Weightless Neural Networks
HoTHP: Hyperbolic Rotary Position Embedding-based Transformer Hawkes Process for Length Extrapolation
Xingu Room
Chair: to be announced
Accuracy, Exposure, and Serendipity in Movie Recommendation
Comparative Evaluation of Local and Quasi-Local Topological Methods for Link Prediction in Complex Networks
Exploring Problem Statements and Source Code Submissions in Online Judges via 2D Embedding Projections
LLM-based Description Enrichment for Short Video Clustering
When Degree Anonymity Is Not Enough: Assessing Structural Re-identification Risk in Social Graphs
Which Feature-Classifier Combinations for Video Sign Language Recognition?
Vila Bela Room
Chapada Room
Chair: Pedro Henrique Silva Souza Barros, Uiversidade Federal de Minas Gerais (UFMG)
Linear Combinations Of Bounded Versus Unbounded Activation Functions: A Comparative Study On Noise Robustness
Auc Is Not Enough: A Utility-Aware Comparison Of Xgboost And Multi-Task Ft Variants For Proxy-Prescriptive Credit Modeling
A Library For Benchmarking Quantum-Circuit Simulators For Quantum Machine Learning
Meta-Learning For Screening Data With Label Noise
A Study Of Action-Sequence Diffusion For Open-Loop Control Under Dry Friction And Stiction
Lightweight Prompt Adaptation Via Self-Organizing Maps
Evaluating Gnn Architecture Embeddings In Conjunction With Self-Supervised Training Objectives For The Session Disambiguation Process
Weighted Metapath Projection For Sir Epidemic Modeling On Tripartite Networks
KeynoteENIAC
Anna Helena Reali Costa, USP
Pantanal Auditorium
Pantanal Auditorium
Andrei Olaru, University Politehnica of Bucharest, Romania
Pantanal Auditorium
Cerrado Auditorium
Chair: Jayr Pereira, Universidade Federal do Cariri (UFCA)
Teaching LLMs Brazilian Healthcare: Injecting Knowledge from Official Clinical Guidelines
RAG-PE: A Hybrid Deterministic and Retrieval-Augmented Architecture for Explainable Multicriteria Regulatory Assessment
CHF-RAG: Coupled Heat-Flow Retrieval-Augmented Generation for Legal Document Retrieval
Mitigating Hallucinations in Tax Guidance: A Structure-Aware Hybrid RAG for Cost-Effective SLM Deployment
Structured Extraction vs Raw Chunks: Retrieval Strategies for Legal Cases
Benchmarking Large Language Models for Structured Information Extraction from Brazilian Occupational Safety Documents
Reinforcement Learning for ENEM-Standard Essay Generation in Brazilian Portuguese with Multi-Competency LLM Judges
Evaluating Location-Aware RAG Approaches for Question Answering over Public Health–Related Content in Brazilian Official Gazettes
Nobres Room
Chair: to be announced
Personalized Federated Fine-Tuning of Language Models
Redes neurais para detecção de séries temporais fora da distribuição em aprendizado de única classe
Chair: to be announced
Sample-Efficient Multi-Task and Multi-Objective Reinforcement Learning by Combining Multiple Behaviors
Araguaia Room
Chair: Fabio Lobato, Universidade de São Paulo (USP)
Understanding Data Augmentation Saturation: A Comparative Study Of Cnns And Vision Transformers For Pollen Grain Classification
Energy-Efficient Horizontal Scaling In Kubernetes With Digital Twins
N-Starw: Normalized Spatiotemporal Random Walks On Dynamic Graphs
Simple Models Or Deep Learning Architectures? A Study On Intergranular Temperature Prediction In Vertical Grain Silos
Br-Rtsd40k: Scaling Brazilian Traffic Sign Recognition Via Semi-Supervised Annotation And Transformer-Based Baselines
Regional Adaptation Of Deep Learning Models For Bird Species Classification In The Pantanal
Optimization Of Deep Learning Models For Bird Identification In The Pantanal Using Embedded Devices
We Will Track You: A Multi-Criterion Evaluation Of Modern Object Detectors For Lizard Behavioral Analysis
Xingu Room
Chair: Daniela Costa; Maucha Gamonal
Do pequeno Asteróide B-612 ao universo: a anotação de O Pequeno Príncipe segundo o modelo Universal Dependencies
Clause Linkage in Brazilian Portuguese Narratives: Evidence from Enhanced Dependency Annotation
Da sintaxe ao discurso: Evidências introdutórias da correlação entre Universal Dependencies e Rhetorical Structure Theory
An Exploratory Study of Discursive Signals Annotation in User-Generated Content
Legal Nugget Extraction for Granular Retrieval over Long Jurisprudential Texts
Vila Bela Room
Chair: George Darmiton da Cunha Cavalcanti, Universidade Federal de Pernambuco (UFPE)
An Empirical Study on the Effects of Missing Labels in Online Continual Learning Alencar
Continual Learning for Sequential Personalization of Small Language Models: A Stability Monitoring Analysis
Assessing Transfer Learning for Turbofan RUL Prediction under Data Scarcity and Domain Shift
Evaluating and Improving Robustness of Once-for-All AutoML approach
Multi-Label Meta-Learning for Pipeline Recommendation: The importance of pipeline diversity
Learning Transferable Instance Hardness: A Meta-Learning Approach to Predict Model Difficulty Across Datasets
Instance-level Hardness Analysis for Multi-label Classification
Target Kernel Design for Alignment-Based Multiple Kernel One-Class Classification
Chapada Room
Chair: to be announced
Beyond Toxicity: Evaluating Language Models for Ableism Detection in Online Social Media
Classification of Descriptive Sufficiency in Public Expenditure Liquidation Records
Integrando Dados Textuais e Estruturados para Compreender Fatores Associados à Infrequência Escolar
Local Anonymization of Free-Text Robbery Police Reports in Brazilian Portuguese: a Safety-Recall-Oriented Multi-Model Benchmark
Risk Assessment of Revictimization in Cases of Domestic Violence Against Women
Where Does Hardness Come From? An Instance Hardness Analysis of Substance-Use Prediction in Brazilian Adolescents
Pantanal Auditorium
Chair: Wandemberg Gibaut, Instituto de Pesquisas Eldorado
JudgeBench-BR: A Replication and Adaptation of a Multi-Domain Benchmark for Evaluating LLM Judges in Brazilian Portuguese
ENADE-QA: A Brazilian Portuguese Dataset for Benchmarking Large Language Models on Health-Related Higher Education Exams
BLUEX v2: Benchmarking LLMs on Open-Ended Questions from Brazilian University Entrance Exams
CAPITU: A Benchmark for Evaluating Instruction-Following in Brazilian Portuguese with Literary Context
MARCA: A Checklist-Based Benchmark for Multilingual Web Search
What Sub-Billion Models Don’t Know About Brazil
Enem-VL: A Multimodal Benchmark of Brazilian Exams for Evaluating Vision-Language Models
How Large Language Models Profile Brazilians: Socioeconomic Attribution and Stereotypes
Cerrado Auditorium
Chair: Josiel Maimone de Figueiredo, UFMT
An Empirical Study Of Simulated Litigation Detection In Brazil: A Machine Learning Approach
Concept-Based Selective Passage Indexing For Neural Ranking In Brazilian Portuguese Legal Information Retrieval
Fact, Legal Thesis, And Noise: Comparing Classical Ml, Bert, And Llms For Triage Of Initial Petitions In Brazilian Portuguese
Small Language Models For Legal Rag In Portuguese: A 4×2 Cross-Domain Evaluation Of Faithfulness, Retrieval Strategies, And Model Scale
Lexdiverse: A Diversity-Aware Data Selection Strategy For Legal Nlp
Can Data Fusion Help Bert Recognize Predatory Litigation Patterns?
Agentic Benchmark Generation And Factorial Evaluation Of Text-To-Sql Architectures On An Industrial Database
Agentic Text-To-Sql In Enterprise Environments: Balancing Planning Paradigms And Pragmatic Self-Repair
Sage-Sql: An Agentic Generator For Synthetic Text-To-Sql Benchmarks Grounded In Real Industrial Schemas
Text-To-Sql Beyond Benchmarks: A Case Study On Large-Scale Deployment
Nobres Room
Chair: to be announced
One Class Learning for Graph Neural Networks
Socially Responsible and Explainable Automated Fact-Checking and Hate Speech Detection
A Comprehensive Exploitation of Instance Selection Methods for Automatic Text Classification: "Doing More with Less"
Araguaia Room
Chair: Alex Marino Gonçalves de Almeida, (Fatec Ourinhos)
Molecular K-Mer Images With Vision Transformers For Blood–Brain Barrier Permeability Prediction
An Explainable Deep Learning Approach For Frontotemporal Dementia Diagnostic Classification From Structural Mri
Clinical Co-Occurrence Priors For Multilabel Classification In Brazilian Gastroscopy Images
An Edge Efficiency Score For Computer Vision Models Applied To Traffic Sign Detection
Efficientnet-Based Myocarditis Detection In T2-Weighted Cmr Images: A Patient-Level Evaluation Strategy
Pseudo-Box Target Sensitivity For Point-Supervised Cervical Cell Detection And Counting
Who Is Who In Judo? Role-Aware Detection In Match Video
An End-To-End Multimodal System For Large-Scale Road Infrastructure Monitoring Using Computer Vision And Geospatial Integration
Image-Based Weight Estimation Of Sheep Using Multitask Convolutional Neural Networks
When Do Spectral-Spatial Transformers Surpass Classical Ensembles In Hyperspectral Image Classification? A Leakage-Aware Supervised Evaluation
Xingu Room
Chair: Thiago Alexandre Salgueiro Pardo, Instituto de Ciências Matemáticas e de Computação da Universidade de São Paulo (ICMC-USP)
Textbook-Enriched Training for Language Models: Boosting Answer Quality in Specialized Contexts
Evaluating Portuguese Tokenizers as Morpheme Sequence in Relation to LLM Downstream Performance
Empirical Study Towards RBAMR-News: a Rule-Based AMR Parser for News in Portuguese
Towards Uniform Meaning Representation for Brazilian Portuguese: Building a First UMR-Annotated Dataset
Validating Semantic Content in Sentence Simplification using OpenIE and Relational Graphs
Text segmentation with mixed data: evaluating symbolic and neural approaches
A Comparative Framework for Diachronic Lexical Semantic Change in Brazilian Portuguese Political Discourse
Vila Bela Room
Chair: Pedro Clarindo da Silva Neto, Instituto Federal de Mato Grosso (IFMT)
Glaucoma Screening Using Transfer Learning with a ConvNeXt V2 Architecture
Synthetic Data Augmentation for Plasmocyte Detection in Bone Marrow Images
Breast Cancer Detection Using Thermal Images via Transfer Learning and Classifier Ensembles
MultiSegUS: A Web-Based System for Multimodel Breast Lesion Segmentation in Ultrasound Images
Enhancing Oral Leukoplakia Segmentation in Histopathological Images Through Heterogeneous Deep Learning Ensemble with Soft Voting
Cascade-Based Machine Learning for Skeletal Muscle and Adipose Tissue Segmentation in CT Images with Coarse Annotations
A Cascaded Multimodal Deep Learning Framework for PET/CT Lesion Segmentation with Metabolic Screening and Dual-Encoder Networks
Beyond Single-Model Segmentation: Consensus-Based Model Selection for Handling Manifestation Variability
Chapada Room
Chair: to be announced
A Generative Model-Driven Pipeline for Post-Hoc Prioritization of Candidates in Nonlinear Optical Materials Discovery
A Hybrid Clustering-Based Approach for Renewal Opportunity Prioritization in Payroll-Deducted Loan Portfolios
Lightweight SPIT Detection Under Adversarial Mimicry
PLC-Embedded Virtual Pressure Sensor with Continual Learning for Mining Tailings Pumping
Residual Hybrid Learning for Data-Limited Formula 1 Lap Time Prediction Across Multiple Circuits
Traditional Machine Learning versus Large Language Models for Predicting Ship Turnaround Times at Brazilian Ports
No activities for the selected events on this day.
Pantanal Auditorium
Towards a Noisy Estimation of Distribution Algorithm for Expensive Binary Optimization Problems
Two Large-Scale Oil and Gas Datasets for Multi-Output Regression Tasks
Beyond Specialized Models: Assessing Vision-Language Models (VLMs) for Zero-Shot Malaria Detection
Explainable Artificial Intelligence for Prediction of Li-Fraumeni Syndrome TP53 p.R337H Founder Variant in Brazil
Semantic Segmentation of Roads and Buildings in Historical Maps from the Dutch Colonial Period
SkinBot-RAG: A Multimodal RAG Framework for Skin Lesion Analysis Using Large Vision-Language Models
SCA-Lex: A Lightweight Forensic Triage Framework based on Few-Shot Semantic Centroid Alignment
Markerless Approach to Gait Analysis in Amputees
Automated Payment Verification of Legal Documents
Laboratory-Based Evaluation of PU Learning for Low-Cost Health Screening: Detecting Hidden Cases of Prediabetes and Diabetes from Self-Reported Data
Quantum Generative Models for Nuclear Engineering: A Comparative Analysis of QGAN and QCBM across Fission, Thermal, and Resonance Spectra
What Matters in RAG for Small Language Models? A Large-Scale Sensitivity Study
A Hybrid Method for Generating Local Rule-Based Explanations for Skin Lesion Images Classification
Credit Assignment in Reinforcement Learning with Group Relative Policy Optimization
Translating Leet Speak: Evaluating PEFT Methods for Text Normalization in Portuguese
Efficient Text Graph Classification via Graph Condensation
CoronaryAI: A Modular and Auditable Pipeline for Automated Extraction of SYNTAX Score Components from Coronary Angiography
New Intent Discovery in Peer-Review Critiques: A Morphosyntactic Approach for Interpretable Cluster Labeling
DyFO: Heterogeneous Temporal Graph Attention with Typed Edge Conditioning for Financial Co-movement Forecasting
Global Deep-Tree Ensembles with Leaf Embeddings for Hierarchical Classification of Transposable Elements
Solving Tetris through Double Deep Q-Learning using Afterstates and a Variable Reward Function
Detection of Symptomatic Regions in Coffee Plantations Using Drone Imagery
Editorial Faithfulness in Reasoning Models: Susceptibility to Authority, Framing, and Information Legitimacy Cues
What Do You Really Mean? Automatically Simplifying Sarcasm and Lexical Ambiguity to Enhance Text Accessibility
LLM-Lobotomy: A Representation Engineering Framework for LLM Behavioral Steering
Exploring Transformers for Predicting the Remaining Useful Life of Turbofan Engines
A Sparsity Protocol for Evaluating Review-Aware Recommendation under Cold-Start Conditions
Interactive Quantum Regressor based on Open Quantum Systems
StereoSet-PT: The stereotype evaluation framework for Brazilian Portuguese language models
Spatio-Temporal Anomaly Detection Using Log-Gaussian Cox Processes with Variational Inference
Survival Prediction in NSCLC: A Comparison of Unimodal and Multimodal Models
Reproducing the EMPT Pipeline: An Open-Source Approach to Motor-Imagery EEG Classification
A Detector and Vision-Language Model Pipeline for Object Description in Maritime Search and Rescue
Weibull Proportional Hazards Model with Deep Learning for Time-Series Memory Failure Prediction
An Automated Framework for Generating Image-Text Datasets from PDF documents
Linear Domain Adaptation for Robust Fault Diagnosis in Electrical Machines
Semantic Segmentation of Coffee Fields Under Spatial Generalization: An Ablation Study with PlanetScope and Sentinel-2
A Data Science-Based, Tactical Role Independent, Comprehensive Metric to Evaluate Professional Football Players
Transformer-based End-to-End Model for Monocular Visual Odometry on Urban Traffic Scenes
A Curriculum-Based Course Timetabling with a focus on optimizing the ENADE score for teaching staff
Dynamic Model Selection for Solar Irradiance Forecasting Using Random Patches Ensembles
Evolutionary Approaches for Parameter Estimation in ODE-based Gene Regulatory Network Models
Back to counting: Conditional Space Graph Representation (CSGR)
Conditional Variational Autoencoder with Rock-Type Awareness for Imbalanced Well-Log Data Augmentation
Video Transformers for Motor Stereotypies Classification in Children with Autism Spectrum Disorder
Comparative Analysis of Convolutional Neural Networks for Personal Protective Equipment (PPE) Detection in Drilling Sites in Mining
Information-Theoretic Instance Hardness: A Minimum Description Length Framework
Agentic Text-to-SQL with Table-Level Retrieval for Protective Relay Settings
The Role of Contextual Grounding in Geoscience Benchmarks
Anatomical Mask Conditioning for Low-Dose CT Denoising: An Architecture-Focused Investigation
Complexity Profiles of Deep Representations: A Layer-wise Meta-Feature Analysis
Adaptive Multi-Objective Reinforcement Learning for Context-Conditioned Water Temperature Control
When Does Multimodal Deep Learning Help Crop Yield Prediction? A Dimensional-Holdout Study on Public Data
Automatic Estimation of Settleable Solids in Biofloc Systems Using Computer Vision Applied to Imhoff Cone Images
Active Visual Inspection with Deep Reinforcement Learning for Disease Diagnosis in Tomato Plants
Semantic Pareto-DQN: A Multi-Objective Reinforcement Learning Framework for Financial Anomaly Detection
Temporal Explainable AI for Open-Set Underwater Acoustic Recognition Using a Band-Specialized Mixture-of-Experts
A Hybrid Exact-Heuristic Framework for the Parallel Machine Total Weighted Tardiness Scheduling Problem
Lightweight CNN for Structural Evolution Detection in Biological Samples via Laser Speckle Pattern Analysis
Automated Public Tender Analysis through a Multi-Stage Pipeline of Classical ML, Transformers, and Large Language Models
Cognitive-Inspired Memory Decay for Collective Mapping in Robot Swarms
WGAN-SDE: Learning Distributions of Stochastic Differential Equation Solutions with Generative Adversarial Networks
Inconsistent Policy Convergence in Undiscounted Reinforcement Learning
Synaptic Bridges: Connecting Internal Neural Network Layers for Multimodal Information Fusion
2.5D U-Net-EL: A 2.5D U-Net with Early Learning Method Applied to Automatic Kidney Segmentation
Wind Turbine Parameter Identification through Kolmogorov-Arnold Networks and Symbolic Distillation
Beyond Sequence: Propagating Inter-Protein Structural Context for Protein Language Models
Probabilistic Geometry of Off-Policy CoT Distillation
Named entity resolution: benchmarking and entity matching strategies
OptiAgent: End-to-End Optimization Modeling via Multi-Agent Iterative Refinement
Structuring Intelligence: Agent-Based Decomposition for Tabular Reasoning with Small Language Models
Evaluating Large Language Models for Decision-Making in Agent-Based Urban Mobility Simulations
Large-scale NISQ circuit fidelity prediction using Graph Transformer to explore the technological frontier
A Neutral and Evidence-Grounded Evaluation of Chunking Strategies in Industrial RAG Systems
Multi-Variety Domain Adaptation for Bidirectional Aymara–Spanish Neural Machine Translation
Hyperparameter-Optimized Heterogeneous Ensembles for Surrogate Modeling of Darcy Flow
Explainable Decision Intelligence for Multi-Supplier Procurement
Extending Human Memory via XR: A Comparative Study of Biological vs. AI-Augmented Recall using Ray-Ban Meta Smart Glasses
Evaluating Clinical Risk Classification Methods in Healthcare Conversations
Does Context Matter? Evaluating Contextual Enrichment in Text-to-SQL for Open Government Data
Cerrado Auditorium
Marcelo Turine, UFMS
Ricardo M. Marcacini, USP
Solange Rezende, USP
Cerrado Auditorium
Juliana Crepalde, VP do Instituto Kunumi
Renata De Paris, Engenheira de Pesquisa
Cerrado Auditorium
Nobres Room
Chair: Carolina Gil Marcelino, Universidade Federal do Rio de Janeiro (UFRJ)
C-DEEPSO-Powell: A Hybrid Approach for Large-Scale Optimization Problems
A Multi-Objective Evolutionary Wrapper to Balance Predictive Power and Cognitive Load in Financial Domains
Learnable Wavelet Transform with Quadrature Mirror Filters for Buy-Sell Trend Classification in the Brazilian Stock Market
Beyond Correlation: Integrating Process Mining and Causal Inference for Root Cause Analysis
CausalGA: An Evolutionary Framework for Causal Structure Learning
Solomonoff Value-Mixtures and the Free Inference Boundary: A Theoretical Framework for Meta-Reinforcement Learning via Finite Bayesian Mixtures of Q-Agents
A Unified Framework for Benchmarking Machine Learning Methods in Genetic Circuit Design
Efficient Table Structure Recognition via Deep Learning and Lightweight Heuristics
Araguaia Room
Chair: Vládia Célia M. Pinheiro, UNIFOR
Diminishing Returns: Llm Accuracy Vs. Computational Cost On Constrained Hardware
The Impact Of Text Representations On Classification Complexity And Predictive Performance
A Comparative Study Of Bert, Llm-Based, And Llm2vec Embeddings For Text Classification
Toward A Fair Evaluation Protocol For Automatic Prompt Optimization Methods
Evaluating Llm Models For Model-To-Text Transformation: Singletons, Ensembles, And Statistical Analysis
Rhea: An Agentic Evaluation And Monitoring Framework
Formal Proof Generation In Deductive Logic Systems With Open Llms: An Empirical Study On Natural Deduction And Analytical Tableau
Comparative Analysis Of Balancing Techniques With Synthetic Data In The Classification Of Police Reports
Xingu Room
Chair: to be announced
A Brazilian Dataset for Financial Sentiment Analysis: An Active Learning-Based Annotation Framework
A Comparative Evaluation of Embedding Models for Retrieval and Clustering of Brazilian Legislative Amendments
A Composite Metric for Abstractive Summarization Evaluation: Integrating Lexical, Semantic, and Factual Dimensions
A Multi-Parser Benchmark over Android Bug-Report Logs
Semantic Macro-Themes in Sertanejo Universitário Lyrics: Topic Modeling with BERTopic
The Role of Retained-Layer Representations in Data Selection for Compressed Transformer Fine-Tuning
Vila Bela Room
Chair: to be announced
Jarvis: LLM-Assisted Plan and Belief Generation via Artifact Introspection
Intelligent Monitoring of Houseplants Using Embedded AI Agents
Autonomous irrigation: How can intelligent agents contribute to plant care?
Proposal of a hardware architecture for communication between Jason agents and FPGA's
Remedino: Embedded Multi-Agent System on the CHON Platform for Assistive Medication Monitoring
Spec-Driven Multi-Agent Corpus Generation for Over-the-Air Wi-Fi Fuzzing
A Multi-Agent LLM Architecture with Dynamic Cascading for On-Premises, LGPD-Compliant Organizational Assistants
Prof. Dr. Marcelo da Silva Reis, UNICAMP
Vila Bela Room
Chapada Room
Miguel Wanderley, Itaú
Chapada Room
Larissa Sayuri Moreira Sugai, Cornell Lab of Ornithology
Carolline Zatta Fieker, CO.BRA; IPBC/INPP
Marinez Isaac Marques, CO.BRA; IPBC/INPP
Pantanal Auditorium
Pantanal Auditorium
No activities in this room at this time.
Cerrado Auditorium
Gabriella Nóbrega, HardwareBR
Cerrado Auditorium
Nobres Room
Chair: Edward Hermann Haeusler, Pontifícia Universidade Católica do Rio de Janeiro (PUC-Rio)
Towards Operationally Robust Deepfake Audio Detection: Synergizing Handcrafted and Self-Supervised Representations
Fairness and Explainability in Portuguese Audio Deepfake Detection: The Effect of Audio Codecs
Automatic Spoken Word Recognition for Very Low-Resource Languages: A Comparison between Zero-Shot and Siamese Approaches in Two Indigenous Languages
Synthetic Data Augmentation for Domain-Specific ASR: Analyzing Proportion Effects Across Pretrained Models
Automatic Speech Recognition and Generation for Brazilian Indigenous Languages: Bororo, Paresi, Enawene Nawe and Xavante
Pseudoword Assessment in Brazilian Portuguese Reading Fluency Tests
LatinX: Aligning a Multilingual TTS Model with Direct Preference Optimization
Neural Speech Enhancement for Conversational AI over Narrowband Telephony: A Benchmark in Brazilian Portuguese
Araguaia Room
Chair: Roney Lira de Sales Santos, UFBA
A Hybrid Rag Pipeline For Linking Fraud News To Public Procurement Records
Benchmarking Text Classifiers For Economic News In Brazilian Portuguese: In-Domain Quality And Out-Of-Distribution Generalization
Stereotype Detection In Llm-Generated Text In Portuguese
High In-Distribution Performance, Fragile Generalization: Generalization Gaps And Shortcut Learning In Brazilian Portuguese Fake News Detection
When The Split Matters: Reorganizing Micro-Edited Pairs For Portuguese Pun Detection
Culturaquant: Quantization And Brazilian Cultural Knowledge In Small Language Models
Mitigating Mina-Br Unbalanced Data Using Uncensored Llm For Hate Speech Detection In Brazilian Portuguese
Xingu Room
Chair: to be announced
Beyond Retrieval: Bidirectional Neuro-Symbolic Validation for Reliable Graph-RAG
Evaluation of LLMs in Answering Questions: A Case Study on POSCOMP
In-Context Teacher–Student Guidance for Open-Weight Browser Agents
KG-RAG: A Hybrid Retrieval Architecture for Institutional Normative Documents
Ontology-Enriched GraphRAG: A Case Study in Portuguese Geoscience Question Answering
Text to SPARQL via LLMs Fine-Tuned with Rewards and Reasoning
Vila Bela Room
Chair: Daniela Costa; Maucha Gamonal
Quando o "…" importa: análise comparada de modalidades de transcrição de hesitações em entrevistas do Corpus Roda Viva
A Comparative Analysis of Adjectives in Image Descriptions in Brazilian Portuguese and English
Classificação automática de sentenças faladas com arquitetura híbrida: análise de estruturas topicalizadas, anti-topicalizadas e canônicas
Inferência Textual em Português Zero-Shot: Especializacão Linguística vs. Raciocínio Explícito em Modelos de Linguagem Compactos
Chapada Room
Chair: to be announced
Integration of LLM-based Intelligent Agents to Assist Decision-Making in Autonomous Vehicles
Automatic Generation of Study Texts from Lecture Slides
Zero-Shot Product Title Normalization with LLMs: A Multi-Model Evaluation
A Comparative Study Between Qwen 3.6 Flash and Small Language Models for Email Prompt Injection in Android Edge Agents
Benchmarking Agentic Capabilities of LLMs in Data Science: Prompt Analysis and Performance of Open-Weight Models
KeynoteBRACIS
Francesca Rossi, IBM
Cerrado Auditorium
Pantanal Auditorium
No activities in this room at this time.
Cerrado Auditorium
No activities in this room at this time.
Nobres Room
Chair: Anderson Meneses, Universidade Federal do Oeste do Pará (UFOPA)
What Shapes Representations in Visual Models? Evidence from Architecture, Training, and Data
Superpixels as an Alternative to Patches in Vision Transformers for Segmentation
Image distortion classification: a joint analysis of convolutional architectures and color models
Multicentric Prototypes Pseudo-Labeling for Source-Free Video Domain Adaptation
Real-Time Fall Detection with CNN-LSTM: Video-Level Evaluation and Cross-Dataset Generalization via Fine-Tuning
Educational Annotation Systems for Computer Vision: Automatic Evaluation and Dataset Generation for Medical Applications
Araguaia Room
Chair: Ricardo M. Marcacini, ICMC/USP
Linguistic Analysis Of Neural-To-Symbolic Translation Failures In Knowledge Graph-Based Fact Checking
Symbolic Ai Versus Llms For Cross-Format Cosmic Functional Size Measurement: A Five-Method Benchmark
Ambiguity-Aware Optimization For Robust Neuro-Symbolic Learning
Explanation-Based Approximate Inference In Stratified Probabilistic Logic Programs
A Grammatical And Semantic Genetic Programming Technique For Symbolic Regression
Evaluation Of Rag Retrieval Strategies With Quantized Qwen3 Models For University Documents
A Resilient Ragops Architecture For Chatbots Over Multimodal Knowledge Bases
Granularity Reciprocal Rank Fusion: Fusing Multiple Chunk Granularities For Robust Retrieval-Augmented Generation
Xingu Room
Chair: to be announced
Beyond Efficiency: The Impact of Instance Selection on Semantic Locality in Transformer-based Text Classification
Classical Tricks for Better Topics: Can Traditional Text Mining Techniques Improve Generative Topic Modeling?
Knowledge Discovery of Emerging Developer Skill Profiles in Telegram Community Using Semantic Embeddings
Monitoring Explanation Drift in Longitudinal NLP Pipelines: A Cross-Domain Study under Temporal Distribution Shift
STAB: A Complexity-Ladder Framework for Evaluating Local LLMs as Autonomous Agents
Vila Bela Room
Chair: Flavia Bernardini, UFF
Subsea Valve Anomaly Detection: A Comparative Analysis Of One-Class Deep Neural Network Architectures}
Detection Of Rotating Looseness Faults In Electric Motors Using Triaxial Vibration Data
Graph-Based Novelty Detection In Passive Sonar
Automated Detection Of Illegal Mining Vessels Using Sar Imagery And Deep Neural Networks
Vision-Language-Action Models: Adaptive And Uncertainty-Gated Policy Distillation
L6 Data Agents: From Individual Intelligence To Collective Autonomy
Simultaneous Optimization Of Wind Farm Layout And Cable Routing Using Multi-Objective Evolutionary Algorithms
Application Of The Lejepa Self-Supervised Learning Technique For The Diagnosis Of Rotating Looseness Faults Using Vibration Data
Chapada Room
Chair: to be announced
Agentic Multi-Document Summarization with Anchoring and Consolidation of Parallel Events: An Experiment with the Four Gospels
BDI-Based Multi-Agent System for Identifying Fraud Patterns in Bidding Processes
Messenger BDI: An Human and Cognitive Agent Communication approach over KQML
EDUPersona: A Multi-Agent Environment for Personalized Guidance through Learning Objects
Design and Implementation of a Multi-Agent Image Preprocessing Framework for Deep Learning
Defining Web Embodiment: Requirements for Embodied Intelligent Agents on the Web
Hugo Valadares, MCTI
Rafael Teixeira Sousa, CEIA/UFG
Nilton Hideki Takagi, UFMT
Pantanal Auditorium
No activities for the selected events on this day.