This book constitutes the post publication of extended paper versions from the 17th International Conference on Agents and Artificial Intelligence, ICAART 2025, which took place in Porto, Portugal, during February 23–25, 2025
The 52 full papers and 39 short papers presented in these two volumes were carefully reviewed and selected from 472 submissions.
The papers are organized in the following topical sections of relevant trends of current research on Agents and Artificial Intelligence, including: Machine Learning, Deep Learning, Multi-Agent Systems, Natural Language Processing, AI and Creativity, Intelligence and Cybersecurity,
Explainable AI, Industrial Applications of AI, Simulation and Agent Models and Architectures.
Table of Contents:
.- Artificial Intelligence.
.- On the Selection of Characteristic Objects in ESP-COMET: A Novel Approach and Assessment Framework.
.- Exploring MFCCs, eGeMAPSv02, and Emobase for Emotion Recognition in Tunisian Dialect.
.- Multicommodity Flow Based Centralized Traffic Routing.
.- Exploration–Exploitation Trade-off in Neural Monte Carlo Tree Search for the Job Shop Scheduling Problem.
.- Exploring the Usability and Applicability of Proof of Learning in Binary Neural Networks.
.- CDRS-PC-JSW: A Context-Driven Collaborative Filtering with Probabilistic Clustering and Divergence-Based Similarity Jensen–Shannon vs Wasserstein.
.- Towards the Selection of Sustainable Suppliers Using Deep Learning.
.- PointTrack: A Weakly-Supervised Segment-and-Track Anything Framework for Microscopy.
.- DGRCL: A Dynamic Contrastive Graph Learning Framework for Modeling Stock Market Evolution.
.- Robust Machine Learning for Harsh Environments: A Framework and Evaluation of Key Architectural Choices.
.- ABBIE-M: Multi-Split Prediction of Sanskrit Compounds by Attention-Based Bi-Encoders.
.- Non-Additive Neural Networks for Classification with Reliability Index.
.- Temporal Fusion Transformers for Autonomous Driving.
.- Uncovering Hidden Biases in Synthetic Java Vulnerability Data for LLM-Based Detection.
.- Multilingual and Sentiment-Aware AI for Toxic Speech Moderation in Political Contexts: An Extended Study of Valencian Legislative Discourse.
.- Dynamic and Partially Observable MAPF: An Extended Evaluation Integrating Informed Path Reuse and Context-Aware Planning.
.- Automatic Data Augmentation of CDT, HDT, and PDT Images for Cognitive Screening.
.- Hybrid Reptile Search Algorithm with Levy Flight for Intrusion Detection.
.- Generative Approaches to Emotion Expression in 3D Human Pose Modeling.
.- MicroVision: A Framework for Instance Segmentation of Yeast Cells and Traps in Microstructured Environment.
.- Stock Price Prediction with Heterogeneous Data Fusion.
.- Improving Answer Generation Quality in LLM-Based Employee Coaching Systems: Evaluation of Next-Generation Models for German Customer Service.
.- CLT-Net: Integrating CNN, LSTM, and Transformer for Review Classification in Online Learning Platforms.
.- A Novel Cuckoo Search Optimisation Enhanced Random Key Addressing the Graph Burning Problem in Social Networks.
.- Hybrid Siamese EfficientNet-B2 and Swin Transformer Network for OCT-Based Prediction of Anti-VEGF Response in Diabetic Macular Edema.
.- Dimensionality Reduction on the SPD Manifold: A Comprehensive Comparative Study Between Conventional and Geometry-Aware Methods.
.- Cellular Automata Applied to the Simulation of Pedestrian Evacuation with Surmountable Obstacles and Route Alternation Through an Impatience Mechanism.
.- Mitigating and Investigating Algorithm Aversion in Medical AI: The Role of Data Labeling and AI Output Adjustment in Diaphragm Detection.
.- Hybrid Transfer-Learning with Custom Centering (TLCC) Framework for Industrial Images Clustering.
.- The Impact of Masking and Multiple Autoencoders on Hierarchically Gated Experts.
.- A Unified Framework and Benchmark Study for Optimization of Diverse, Plausible, and Actionable Counterfactual Explanations.
.- Graph-TransformerMultimodalLearningforMulti-AgentTrajectory ForecastingusingMOPSO-OptimizedTemporalKolmogorov-Arnold Networks.
.- A ResearchAgendaforUsabilityandGeneralisationinReinforcement
Learning.
