---
dataset_info:
  features:
  - name: a
    dtype: string
  - name: b
    dtype: string
  - name: label
    dtype: int64
  splits:
  - name: train
    num_bytes: 685896937
    num_examples: 181000
  - name: valid
    num_bytes: 17346151
    num_examples: 4584
  - name: test
    num_bytes: 2872780
    num_examples: 753
  download_size: 208705249
  dataset_size: 706115868
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 CVD in the paper [Contrastive Learning and Mixture of Experts Enables Precise Vector Embeddings](arxiv.org/abs/2401.15713)