Research
Compositionality & Capabilities of Language Models
- Measuring Idiomaticity in Text Embedding Models with 𝜀-compositionality
- EACL 2026 paper
- Systematic Generalization in Language Models Scales with Information Entropy
- ACL 2025 findings paper
- Compositional Generalization with Grounded Language Models
- ACL 2024 findings paper
Information Extraction
- Abstractive Event Analysis of Armed Conflicts: Introducing the UCDP-AEC Dataset
- Generative Approaches to Event Extraction: Survey and Outlook
- Socio-political Events of Conflict and Unrest: A Survey of Available Datasets
- Deep Learning for Unsupervised Relation Extraction
- My PhD thesis, on which I worked very (too?) hard to make it more than a simple aggregation of papers.
- Unsupervised Information Extraction: Regularizing Discriminative Approaches with Relation Distribution Losses
- ACL 2019 long paper corresponding to the first half of my thesis.
Other
- Fine-tuning and Sampling Strategies for Multimodal Role Labeling of Entities under Class Imbalance
- CONSTRAINT 2022 paper on a thesis-unrelated topic because I wanted to colaborate with other people during my PhD candidacy.
- Neural Machine Translation with Memory Network Based Attention
- My end of master's degree internship. The goal was to use a memory network decoder in an NMT system.
- Kaggle: Taxi Destination Prediction
- First place in a competition for the 2015 ECML PKDD conference. The goal was to predict the destination of a taxi given a prefix of its trajectory.