不正な原因
コレクションでコンテンツを整理
必要に応じて、コンテンツの保存と分類を行います。
間違った原因
因果推論は人間の知性の特徴の 1 つです。
Corr2cause は 400,000 サンプルを超える大規模なデータセットであり、関連する論文では 17 の既存の LLM が評価されています。
全体として、Corr2cause には 415,944 個のサンプルが含まれており、そのうち 18.57% が有効なサンプルです。前提の平均長は 424.11 トークン、仮説は 10.83 トークンです。データは、それぞれ 411,452 個のトレーニング サンプル、2,246 個の開発サンプルとテスト サンプルに分割されます。データセットの主な目的は LLM のパフォーマンスのベンチマークであるため、テストおよび開発セットは、あらゆるサイズのグラフを包括的にカバーするように優先されています。
スプリット | 例 |
---|
'dev' | 2,246 |
'test' | 2,246 |
'train' | 411,452 |
FeaturesDict({
'input': Text(shape=(), dtype=string),
'label': int64,
})
特徴 | クラス | 形 | Dタイプ | 説明 |
---|
| 特徴辞書 | | | |
入力 | 文章 | | 弦 | |
ラベル | テンソル | | int64 | |
@misc{jin2023large,
title={Can Large Language Models Infer Causation from Correlation?},
author={Zhijing Jin and Jiarui Liu and Zhiheng Lyu and Spencer Poff and Mrinmaya Sachan and Rada Mihalcea and Mona Diab and Bernhard Schölkopf},
year={2023},
eprint={2306.05836},
archivePrefix={arXiv},
primaryClass={cs.CL}
}
特に記載のない限り、このページのコンテンツはクリエイティブ・コモンズの表示 4.0 ライセンスにより使用許諾されます。コードサンプルは Apache 2.0 ライセンスにより使用許諾されます。詳しくは、Google Developers サイトのポリシーをご覧ください。Java は Oracle および関連会社の登録商標です。
最終更新日 2023-09-09 UTC。
[null,null,["最終更新日 2023-09-09 UTC。"],[],[],null,["# corr2cause\n\n\u003cbr /\u003e\n\n- **Description**:\n\nCorr2cause\n==========\n\nCausal inference is one of the hallmarks of human intelligence.\n\nCorr2cause is a large-scale dataset of more than 400K samples, on which\nseventeen existing LLMs are evaluated in the related paper.\n\nOverall, Corr2cause contains 415,944 samples, with 18.57% in valid samples. The\naverage length of the premise is 424.11 tokens, and hypothesis 10.83 tokens. The\ndata is split into 411,452 training samples, 2,246 development and test samples,\nrespectively. Since the main purpose of the dataset is to benchmark the\nperformance of LLMs, the test and development sets have been prioritized to have\na comprehensive coverage over all sizes of graphs.\n\n- **Homepage** :\n \u003chttps://github.com/causalNLP/corr2cause/tree/main\u003e\n\n- **Source code** :\n [`tfds.datasets.corr2cause.Builder`](https://github.com/tensorflow/datasets/tree/master/tensorflow_datasets/datasets/corr2cause/corr2cause_dataset_builder.py)\n\n- **Versions**:\n\n - **`1.0.0`** (default): Initial release.\n- **Download size** : `727.22 MiB`\n\n- **Dataset size** : `739.91 MiB`\n\n- **Auto-cached**\n ([documentation](https://www.tensorflow.org/datasets/performances#auto-caching)):\n No\n\n- **Splits**:\n\n| Split | Examples |\n|-----------|----------|\n| `'dev'` | 2,246 |\n| `'test'` | 2,246 |\n| `'train'` | 411,452 |\n\n- **Feature structure**:\n\n FeaturesDict({\n 'input': Text(shape=(), dtype=string),\n 'label': int64,\n })\n\n- **Feature documentation**:\n\n| Feature | Class | Shape | Dtype | Description |\n|---------|--------------|-------|--------|-------------|\n| | FeaturesDict | | | |\n| input | Text | | string | |\n| label | Tensor | | int64 | |\n\n- **Supervised keys** (See\n [`as_supervised` doc](https://www.tensorflow.org/datasets/api_docs/python/tfds/load#args)):\n `None`\n\n- **Figure**\n ([tfds.show_examples](https://www.tensorflow.org/datasets/api_docs/python/tfds/visualization/show_examples)):\n Not supported.\n\n- **Examples**\n ([tfds.as_dataframe](https://www.tensorflow.org/datasets/api_docs/python/tfds/as_dataframe)):\n\nDisplay examples... \n\n- **Citation**:\n\n @misc{jin2023large,\n title={Can Large Language Models Infer Causation from Correlation?},\n author={Zhijing Jin and Jiarui Liu and Zhiheng Lyu and Spencer Poff and Mrinmaya Sachan and Rada Mihalcea and Mona Diab and Bernhard Schölkopf},\n year={2023},\n eprint={2306.05836},\n archivePrefix={arXiv},\n primaryClass={cs.CL}\n }"]]