Scholar
Gabriele Prato
Google Scholar ID: 9D_y8ssAAAAJ
Mila, Université de Montréal
Citations & Impact
Citations
138
H-index
4
i10-index
2
Publications
9
Co-authors
13
list available
Publications
Resume (English only)
Academic Achievements
- Do Large Language Models Know How Much They Know?, EMNLP 2024
- EpiK-Eval: Evaluation for Language Models as Epistemic Models, EMNLP 2023
- PatchBlender: A Motion Prior for Video Transformers, NeurIPS 2022 Workshop
- Scaling Laws for the Few-Shot Adaptation of Pre-trained Image Classifiers, ICML 2021 Workshop
- Fully Quantized Transformer for Machine Translation, Findings of EMNLP 2020
- Towards Lossless Encoding of Sentences, ACL 2019
- Blog Post: EpiK-Eval and the Role of Knowledge Consolidation in Language Models, December 22, 2023
Background
- Final-year PhD candidate at Mila, University of Montreal, expecting to graduate in spring or summer 2025
- Research focuses on fundamental aspects of Large Language Models (LLMs)
- Explores how data segmentation affects critical functions such as parametric knowledge retrieval and latent multi-hop reasoning
- Aims to address inherent limitations of LLMs by developing methods for knowledge consolidation during training
- Committed to open-ended research and academic exploration to produce impactful, publishable work
- Seeking research-focused industry roles aligned with these values
