About me
I completed my PhD in Artificial Intelligence and NLP at Sapienza University of Rome, advised by Roberto Navigli in the Sapienza NLP research group. My research focuses on Multilinguality and LLM Evaluation, with a particular emphasis on Machine Translation.
In 2025-2026, I completed internships at Microsoft and Amazon, working on Machine Translation evaluation and Process Reward Modeling for Tool-Calling LLM Agents. Since 2025, I have also served as a WMT shared tasks organizer.
News
I completed my Amazon Applied Scientist Internship in Paris, where I worked on Process Reward Modeling for Tool-Calling LLM Agents.
Our paper PEAR: Pairwise Evaluation for Automatic Relative Scoring in Machine Translation, resulting from my Microsoft Research Internship, has been accepted at the ACL 2026 Main Conference.
I gave an invited talk titled Towards More Informative MT Evaluation at Meta FAIR in Paris.
I started co-leading the Segment-Level Quality Score Prediction task at the WMT Automated Translation Quality Shared Task.
I successfully defended my PhD thesis, Towards Better Measurement of Progress in Machine Translation: Evaluation and Meta-Evaluation.
I started my Amazon Applied Scientist Internship in Paris, working with Bruno Charron, Jayasimha Talur, and Saab Mansour.
Our shared-task paper Findings of the WMT25 General Machine Translation Shared Task: Time to Stop Evaluating on Easy Test Sets appeared at WMT 2025.
Our shared-task paper Findings of the WMT25 Multilingual Instruction Shared Task: Persistent Hurdles in Reasoning, Generation, and Evaluation appeared at WMT 2025.
I completed my Microsoft Research Internship in Redmond, where I worked on Machine Translation evaluation.
Our paper Estimating Machine Translation Difficulty has been accepted to Findings of EMNLP 2025.
I gave an invited talk at Google’s Multilinguality Reading Group presenting our paper Has Machine Translation Evaluation Achieved Human Parity? The Human Reference and the Limits of Progress.
Our paper Has Machine Translation Evaluation Achieved Human Parity? The Human Reference and the Limits of Progress received a SAC Highlights Award at ACL 2025.
I started my Microsoft Research Internship at Microsoft’s headquarters in Redmond, working with Roman Grundkiewicz and Matt Post.
I completed my Visiting PhD Researcher period at the University of Amsterdam.
Our paper Has Machine Translation Evaluation Achieved Human Parity? The Human Reference and the Limits of Progress has been accepted at the ACL 2025 Main Conference.
I started serving as a WMT Shared Task organizer for the General Machine Translation Shared Task and the WMT25 Multilingual Instruction Shared Task.
I started my Visiting PhD Researcher period at the University of Amsterdam, working with Raquel Fernández in the Dialogue Modelling Group.
I gave an invited talk titled The Opacity in MT Evaluation and Meta-Evaluation at Microsoft Translator Research.
Our paper Beyond Correlation: Interpretable Evaluation of Machine Translation Metrics has been accepted at the EMNLP 2024 Main Conference.
Our paper Guardians of the Machine Translation Meta-Evaluation: Sentinel Metrics Fall In! has been accepted at the ACL 2024 Main Conference.
Our paper Analyzing Homonymy Disambiguation Capabilities of Pretrained Language Models has been accepted at LREC-COLING 2024.
I started my PhD in Artificial Intelligence at Sapienza University of Rome, advised by Roberto Navigli in the Sapienza NLP research group.
Our paper MaTESe: Machine Translation Evaluation as a Sequence Tagging Problem has been accepted at WMT 2022.