---
dataset_info:
  features:
  - name: a
    dtype: string
  - name: b
    dtype: string
  - name: label
    dtype: int64
  splits:
  - name: train
    num_bytes: 477301584
    num_examples: 132453
  - name: valid
    num_bytes: 9596971
    num_examples: 2676
  - name: test
    num_bytes: 4758204
    num_examples: 1294
  download_size: 200765538
  dataset_size: 491656759
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
  - split: test
    path: data/test-*
---
This dataset contains the cocitation abstracts related to COPD in the paper [Contrastive Learning and Mixture of Experts Enables Precise Vector Embeddings](arxiv.org/abs/2401.15713)