Program – ReSAISE 2025
October 21, 2025 – São Paulo, Brazil (UTC−3)
Session I - Trustworthy AI: Security, Testing, and Reliability (14:00 – 15:30)
- Detecting Stealthy Data Poisoning Attacks in AI Code Generators
Cristina Improta
- An Evaluation Framework for Software Unit Tests Generated by Large Language Models
Steven Thanthan Nguyen, Smita Bharath, Hajar Homayouni
- On the Adaptability of ML Reliability Estimation Techniques to Federated Settings
Junjian Yan, Paulo Carvalho, Jorge Henriques, João Loureiro, Chan-Tong Lam, Henrique Madeira
- AnchorChain: Immutable Referencing of AI Memory States through Blockchain-Indexed Volumetric Data Linking
Craig Wright
- Anomaly Detection Techniques for Enhanced Open Radio Access Network Cybersecurity
Francisco Neto, Paulo da Silva, Erika Alves, Thomas Paiva, João Labres, Ramon Carvalho
Coffee Break (15:30 – 16:00)
Session II - Causal Insights and Dependability in AI Systems(16:00 – 17:30)
- KEYNOTE: Towards an Interpretable Science of Deep Learning for Secure Software Engineering: A Causal Inference View
Denys Poshyvanyk
- ChaosLLM: A Dependability Testing Approach for Tool-calling Agents
Antonio Ken Iannillo
- QuickCommand: A Low-Latency NLP Pipeline for Reliable UAV Telepresence
Mohamed El-Hadedy, Benny Cheng, Wen-Mei Hwu