Davide Baldelli
davide.baldelli@mila.quebec - davide.baldelli@etud.polymtl.ca
Hi, I’m Davide 😊. I am a first-year PhD student at MILA and Polytechnique Montréal, advised by Sarath Chandar and Amal Zouaq.
I first worked on transformer-based information retrieval and on foundation models for computer-aided design. I now focus on language-model-based agents that use working memory or a private scratchpad, and on the faithfulness of language models’ self-explanations. My goal is to develop agents that can keep private state, reason over it, and produce explanations that accurately reflect the underlying model processes.
Previously, I was a researcher at the National Institute of Informatics (Tokyo) under the supervision of Akiko Aizawa. I received an MSc in Artificial Intelligence from the University of Bologna in October 2023, where I was supervised by Paolo Torroni.
news
| Aug 15, 2025 | I have been selected to join ARENA 7.0 (Alignment Research Engineer Accelerator), which will be held at LISA (London Initiative for Safe AI), London, UK, in January 2026. |
|---|---|
| Jul 13, 2025 | Our paper “CADmium: Fine-Tuning Code Language Models for Text-Driven Sequential CAD Design” is now on arXiv: arXiv:2507.09792. |
| Jun 10, 2025 | I was selected to participate in the Y Combinator AI Startup School in San Francisco. |
| Jan 15, 2025 | I started my PhD at MILA and Polytechnique Montréal with Sarath Chandar and Amal Zouaq. |
| Dec 10, 2023 | Our paper “TWOLAR: a TWO-step LLM-Augmented distillation method for passage Reranking” has been accepted at ECIR 2024! |
| Oct 01, 2023 | I obtained the MSc in AI with maximum grade from the University of Bologna. |
