<aside> 🔥

Interested? got an idea? send an email to [email protected]

</aside>

Keep an eye on the projects. They like to change.

Things I’ve been working on: https://scholar.google.it/citations?user=1xd52jMAAAAJ&hl=en

Sections

• Project List
• Main Collaborations
• Past Thesis

Project List

• Use OpenAI Gymnasium to design environments for studying LLM behavior and decision-making in planning tasks, analyzing strategies and performance. cfr. https://dl.acm.org/doi/abs/10.1145/3586183.3606763.
• Adapt AlphaGo Zero-style models to tackle LLM challenges, focusing on improving complex reasoning paths and automatic prompt optimization using self-play and reinforcement learning. cfr https://arxiv.org/pdf/1712.01815, https://www.nature.com/articles/nature16961, https://www.nature.com/articles/s41586-021-03819-2.
• Apply LLMs to Information Retrieval (IR) problems, evaluating their effectiveness in query understanding, document ranking, and relevance prediction to improve retrieval performance and user satisfaction.
• Use LLMs and optimization techniques to generate novel problems, emphasizing creativity and diversity in formulations for applications in combinatorial optimization.
• Engineer LLM-based real-time conversational agents for healthcare applications, incorporating sentiment analysis and adaptive tone adjustments to enhance patient interactions and support.
• Use synthetic datasets to address biases in LLMs, ensuring fairer outcomes in applications like hiring, lending, or healthcare.
• Extend LLM capabilities to process and integrate data from multiple modalities, such as text, images, and audio, for comprehensive understanding.
• Study the robustness and social behavior of LLMs by using questionnaires to simulate social interactions, evaluating their consistency, adaptability, and alignment with human expectations. cfr https://dl.acm.org/doi/abs/10.1145/3627673.3680002.
• Generate synthetic datasets to align LLMs for specific tasks, using strategies like Proximal Policy Optimization (PPO) to improve performance and adaptability. cfr https://arxiv.org/abs/2406.08464, https://arxiv.org/abs/2407.16216.
• Use agent-based systems to solve complex tasks and simulate intricate environments, exploring emergent behavior and collaborative strategies in dynamic settings. cfr https://dl.acm.org/doi/abs/10.1145/3586183.3606763.