README.md

---
task_categories:
- text-classification
language:
- en
tags:
- natural-language-inference
- nli
- propositions
pretty_name: atomic-snli
configs:
- config_name: default
  data_files:
  - split: train
    path: train.jsonl
  - split: validation
    path: validation.jsonl
  - split: test
    path: test.jsonl
---

# atomic-snli

Atomic propositions for the premise and hypothesis of each NLI pair, derived from
[`stanfordnlp/snli`](https://huggingface.co/datasets/stanfordnlp/snli). Each sentence was decomposed into standalone atomic propositions;
these propositions are joined back to the original NLI pairs.

## Columns

| column | type | description |
|---|---|---|
| `premise` | string | original premise sentence |
| `hypothesis` | string | original hypothesis sentence |
| `label` | int | 0 = entailment, 1 = neutral, 2 = contradiction |
| `premise_propositions` | list[string] | atomic propositions of the premise |
| `hypothesis_propositions` | list[string] | atomic propositions of the hypothesis |

When the decomposition produced nothing for a sentence, the full sentence is kept as the
single proposition (no empty lists).

## Splits

| split | examples |
|---|---|
| `train` | 549367 |
| `validation` | 9842 |
| `test` | 9824 |

## Usage

```python
from datasets import load_dataset
ds = load_dataset("Younes2E/atomic-snli")
```

## Source

Derived from [`stanfordnlp/snli`](https://huggingface.co/datasets/stanfordnlp/snli). See the original dataset for its license and terms.