nli_tr_for_simcse / README.md
emrecan's picture
create readme
2997843
|
raw
history blame
728 Bytes
metadata
languages:
  - tr
size_categories:
  - 100K<n<1M
source_datasets:
  - nli_tr
task_categories:
  - text-scoring
task_ids:
  - semantic-similarity-scoring

NLI-TR for Supervised SimCSE

This dataset is a modified version of NLI-TR dataset. Its intended use is to train Supervised SimCSE models for sentence-embeddings. Steps followed to produce this dataset are listed below:

  1. Merge train split of snli_tr and multinli_tr subsets.
  2. Find every premise that has an entailment hypothesis and a contradiction hypothesis.
  3. Write found triplets into sent0 (premise), sent1 (entailment hypothesis), hard_neg (contradiction hypothesis) format.