Paper: Lexicon Learning for Few-Shot Neural Sequence Modeling
MIT License
Paper: Learning to Recombine and Resample Data for Compositional Generalization
Code to reproduce experiments from the paper "Continual Pre-Training Mitigates Forgetting in Lang...
Influence Experiments
Black-box language model explanation by context length probing
The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑🔬
BAD-VAE: A VAE framework for unsupervised disentanglement of sequential data
Neural question generation using transformers
Ift 6759 projects
Implementation of "Analysing Mathematical Reasoning Abilities of Neural Models"