Richard Diehl Martinez

language model (5) small language model (4) named entity recognition (3) language model pretraining (2) few-shot learning (2) representation learning (2) learning dynamics (2) text generation (1) multi-task learning (1) model compression (1) cross-lingual transfer (1) natural language generation (1) attention mechanism (1) self-supervised learning (1) zero-shot learning (1) model architecture (1) model pretraining (1) curriculum learning (1) child-directed speech (1) convergence analysis (1)