B.Sc. Student in Applied Computer Science and Artificial Intelligence
Sapienza University of Rome, Italy
Website · Hugging Face · LinkedIn · Email · Academic CV
I study information retrieval and retrieval-augmented generation, focusing on when improvements in retrieval do—or do not—lead to more accurate and grounded answers.
My work combines controlled retrieval and generation experiments with paired evaluation, failure analysis, versioned manifests, and inspectable artifacts.
| Project | Focus | Current scope |
|---|---|---|
| msmarco-genqa | Retrieval-augmented generation on MS MARCO | Retrieval, reranking, generation, grounding analysis, paired statistical evaluation, and reproducible experiment reports |
| rag-observatory · Live Space · Toy Dataset | Trace-based analysis for RAG systems | Research prototype for inspecting retrieved evidence, generated answers, execution traces, and failure labels |
| CiboCompass | Offline-resilient mobile food exploration | React Native, Go, and SQLite application with local caching, a persistent retry queue, and idempotent submission |
Research question: When does better retrieval improve grounded generation, and when do conventional evaluation metrics hide the failure?
Current study: Controlled retrieve–rerank–generate experiments on MS MARCO, using paired statistical evaluation, explicit grounding measures, and per-example failure analysis.
Research direction: Developing a versioned RAG failure taxonomy that distinguishes retrieval, reranking, evidence-use, and generation errors.