Discourse Based Evaluation of Language Understanding

Code to reproduce experiments from the paper "Continual Pre-Training Mitigates Forgetting in Language and Vision" https://arxiv.org/abs/2205.09357

Fair quantitative comparison of NLP embeddings from GloVe to RoBERTa with Sequential Bayesian Optimization fine-tuning using Flair and SentEval. Extension of HyperOpt library to log_b priors.

Composition of embeddings

Code for "Model Agnostic Knowledge Transfer Methods for Sentence Embedding Models" paper by Gunel and Amasyali.

Advanced Topics in Computational Semantics (ATCS) - Practical 1