Publications | Aleksandar Fontanahttp://gildarts777.github.io/publications/PublicationsHugoBlox Kit (https://hugoblox.com)en-ushttp://gildarts777.github.io/media/icon_hu_da05098ef60dc2e7.pngPublicationshttp://gildarts777.github.io/publications/On the Hidden Objective Biases of Group-based Reinforcement Learninghttp://gildarts777.github.io/publications/fontana-2026-hidden/Thu, 01 Jan 2026 00:00:00 +0000http://gildarts777.github.io/publications/fontana-2026-hidden/GTPO: Stabilizing Group Relative Policy Optimization via Gradient and Entropy Controlhttp://gildarts777.github.io/publications/simoni-2025-gtpo/Wed, 01 Jan 2025 00:00:00 +0000http://gildarts777.github.io/publications/simoni-2025-gtpo/Improving LLM Reasoning for Vulnerability Detection via Group Relative Policy Optimizationhttp://gildarts777.github.io/publications/simoni-2025-improving/Wed, 01 Jan 2025 00:00:00 +0000http://gildarts777.github.io/publications/simoni-2025-improving/TITAN: Graph-Executable Reasoning for Cyber Threat Intelligencehttp://gildarts777.github.io/publications/simoni-2025-titan/Wed, 01 Jan 2025 00:00:00 +0000http://gildarts777.github.io/publications/simoni-2025-titan/Unmasking model behavior: How llms reason on vulnerability detectionhttp://gildarts777.github.io/publications/fontana-2025-unmasking/Wed, 01 Jan 2025 00:00:00 +0000http://gildarts777.github.io/publications/fontana-2025-unmasking/