wl-research / nubia
NUBIA (NeUral Based Interchangeability Assessor) is a new SoTA evaluation metric for text generation
☆52Updated last year
Related projects: ⓘ
- Code for the Shortformer model, from the ACL 2021 paper by Ofir Press, Noah A. Smith and Mike Lewis.☆146Updated 3 years ago
- ☆101Updated 3 years ago
- ☆108Updated last year
- HetSeq: Distributed GPU Training on Heterogeneous Infrastructure☆105Updated last year
- A 🤗-style implementation of BERT using lambda layers instead of self-attention☆70Updated 3 years ago
- The Python library with command line tools to interact with Dynabench(https://dynabench.org/), such as uploading models.☆54Updated 2 years ago
- Babysit your preemptible TPUs☆84Updated last year
- jiant-dev☆28Updated 3 years ago
- Interactive Model Iteration with Weak Supervision and Pre-Trained Embeddings☆76Updated 2 years ago
- A diff tool for language models☆42Updated 8 months ago
- Question-answers, collected from Google☆124Updated 3 years ago
- Evaluation suite for large-scale language models.☆123Updated 3 years ago
- Module 0 - Fundamentals☆95Updated 3 weeks ago
- MILES is a multilingual text simplifier inspired by LSBert - A BERT-based lexical simplification approach proposed in 2018. Unlike LSBert…☆48Updated 3 years ago
- A word2vec negative sampling implementation with correct CBOW update.☆260Updated 2 years ago
- Implementation of Marge, Pre-training via Paraphrasing, in Pytorch☆75Updated 3 years ago
- A set of tools for leveraging pre-trained embeddings, active learning and model explainability for effecient document classification☆29Updated 2 years ago
- Check if you have training samples in your test set☆64Updated 2 years ago
- Code for scaling Transformers☆26Updated 3 years ago
- API for accessing the GraphLog dataset☆90Updated 4 months ago
- ☆38Updated last year
- LM Pretraining with PyTorch/TPU☆131Updated 4 years ago
- Trains Transformer model variants. Data isn't shuffled between batches.☆140Updated last year
- QED: A Framework and Dataset for Explanations in Question Answering☆114Updated 3 years ago
- ☆73Updated 3 years ago
- Python Research Framework☆107Updated last year
- diagNNose is a Python library that facilitates a broad set of tools for analysing hidden activations of neural models.☆81Updated 10 months ago
- Code for obtaining the Curation Corpus abstractive text summarisation dataset☆122Updated 3 years ago
- Create interactive textual heat maps for Jupiter notebooks☆196Updated 3 months ago
- A biologically inspired method to create sparse, binary word vectors☆36Updated 2 years ago