Andre Martins

Long-Context Generalization with Sparse Attention

Should We Still Pretrain Encoders with Masked Language Modeling?

Movie Facts and Fibs (MF$^2$): A Benchmark for Long Movie Understanding

Sample-efficient Integration of New Modalities into Large Language Models

TowerVision : Understanding and Improving Multilinguality in Vision-Language Models

AdaSplash: Adaptive Sparse Flash Attention

M-Prometheus: A Suite of Open Multilingual LLM Judges

EuroBERT: Scaling Multilingual Encoders for European Languages

Do LLMs Understand Your Translations? Evaluating Paragraph-level MT with Question Answering

Multilingual Contextualization of Large Language Models for Document-Level Machine Translation

$\infty$-Video: A Training-Free Approach to Long Video Understanding via Continuous-Time Memory Consolidation

Zero-shot Benchmarking: A Framework for Flexible and Scalable Automatic Evaluation of Language Models