The dataset viewer is not available because its heuristics could not detect any supported data files. You can try uploading some data files, or configuring the data files location manually.
YAML Metadata Warning:The task_categories "text2text-generation" is not in the official list: text-classification, token-classification, table-question-answering, question-answering, zero-shot-classification, translation, summarization, feature-extraction, text-generation, fill-mask, sentence-similarity, text-to-speech, text-to-audio, automatic-speech-recognition, audio-to-audio, audio-classification, audio-text-to-text, voice-activity-detection, depth-estimation, image-classification, object-detection, image-segmentation, text-to-image, image-to-text, image-to-image, image-to-video, unconditional-image-generation, video-classification, reinforcement-learning, robotics, tabular-classification, tabular-regression, tabular-to-text, table-to-text, multiple-choice, text-ranking, text-retrieval, time-series-forecasting, text-to-video, image-text-to-text, image-text-to-image, image-text-to-video, visual-question-answering, document-question-answering, zero-shot-image-classification, graph-ml, mask-generation, zero-shot-object-detection, text-to-3d, image-to-3d, image-feature-extraction, video-text-to-text, keypoint-detection, visual-document-retrieval, any-to-any, video-to-video, other
allenai/atomic(複製)
データの内容
PersonX などを主語にした日常の出来事に対する、if-then の常識推論(結果・反応・意図など)を集めたデータ(ATOMIC)。1 行が 1 件のイベント(event)で、9 つの次元(oEffect、oReact、xIntent など)のそれぞれに、作業者が付けた回答が JSON のリストで入る。イベントはブログ・物語・書籍から自動的に抜き出したもので、暴力的な行動を描いたものも残してある。企業や市場の因果は扱わない。
データの概要
| 項目 | 値 |
|---|---|
| 行数 | 249,747(1 構成・3 分割の合計) |
| 列数 | 12 |
| 容量 | 15.37 KB(ファイル 3 本) |
| ファイルの形式 | py 1 |
| 言語 | en |
| 公開 | public |
構成と分割
| 構成(config) | 分割(split)ごとの行数 | 列数 |
|---|---|---|
atomic |
train 202,271、test 24,856、validation 22,620 |
12 |
列(構成 atomic)
| 列 | 型 |
|---|---|
event |
string |
oEffect |
list<string> |
oReact |
list<string> |
oWant |
list<string> |
xAttr |
list<string> |
xEffect |
list<string> |
xIntent |
list<string> |
xNeed |
list<string> |
xReact |
list<string> |
xWant |
list<string> |
prefix |
list<string> |
split |
string |
行数と列は元のカードの YAML(dataset_info)の記載。データセットビューアでは読めなかった。全データセットの一覧は HYUKI5/dataset-catalog にある。
複製について
allenai/atomic をリビジョン 5cd968cca3ad16bb3323948d038182b1d5f77902 の時点でそのまま複製したもの。中身は変えていない。判断モデル Laya(convaiinnovations/laya)の微調整に使う候補として、調査メモ(Laya 学習用データセット調査)に載せたデータを集めている。
出典
- 元のデータセット: https://huggingface.co/datasets/allenai/atomic
- 複製したリビジョン:
5cd968cca3ad16bb3323948d038182b1d5f77902(2025-01-13 15:13:50+00:00) - 元のライセンス表記:
['cc-by-4.0']
論文
- 元のカードに論文へのリンクは見つからなかった
アノテーションの方法
収集の根拠
本データセットは、常識推論を支援するために収集・作成された。
初期のデータ収集と正規化
詳細は研究論文とウェブサイトを参照されたい。本データセットは、ワシントン大学がクラウドソーシングで集めたデータを用いて作成された。
アノテーションの過程
フォームに沿って人手で行われたアノテーション。
微調整に使うときの注意
- 調査メモの台帳での区分: B(継承条件・gated・モデル由来の条件・上流の規約などが付く)。区分は記載の文言で機械的に付けたもので、法的な判定ではない。学習に入れる前に元のカードとライセンスの原文で確かめる。
- Laya の入力は 2,048 token まで(ModernBERT-large / mmBERT)。長い本文は先頭から黙って切られるので、切れる行は事前に除くか、要約・切り出しをしてから入れる。
- 分割は記事・文書単位で行い、同じ本文から作った別の問いを学習側と較正側に分けない。温度と閾値は学習に使っていない分割で別々に当てる。
調査メモの記述
| メモ | 節 | ライセンス(記載) | 確認 | 注記 |
|---|---|---|---|---|
ontology_kg_entity.md |
Q5. KG トリプルの真偽判定・常識知識・上位下位・時間関係・オントロジー整合のデータにどんなものがあり、Laya に載るか (L157) | allenai/atomic: cc-by-4.0(HF タグ)。Estwld 版: タグなし | 確認済 | 上流/条件注記 / 確認印なし(各ファイルの凡例により確認済扱い) / 原文注記: スクリプト版は datasets 4 以降で読めない可能性。ATOMIC は日常イベント(PersonX…)で、企業・市場の因果は扱わない |
元のカード(日本語訳)
以下は元のデータセットのカードの本文を日本語に訳したもの(機械翻訳)。訳に疑問があるときは、末尾の原文を確かめる。
An Atlas of Machine Commonsense for If-Then Reasoning - Atomic Common Sense Dataset のデータセットカード
目次
- データセットの説明
- データセットの概要
- 対応タスクとリーダーボード
- 言語
- データセットの構造
- データの例
- データのフィールド
- データの分割
- データセットの作成
- 収集の根拠
- 元データ
- アノテーション
- 個人情報と機微な情報
- データの利用に関する考慮事項
- データセットの社会的影響
- バイアスについての議論
- その他の既知の制限
- 追加情報
- データセットのキュレーター
- ライセンス情報
- 引用情報
- 貢献者
データセットの説明
- ホームページ: https://homes.cs.washington.edu/~msap/atomic/
- リポジトリ: https://homes.cs.washington.edu/~msap/atomic/
- 論文: Maarten Sap, Ronan LeBras, Emily Allaway, Chandra Bhagavatula, Nicholas Lourie, Hannah Rashkin, Brendan Roof, Noah A. Smith & Yejin Choi (2019). ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning. AAAI
データセットの概要
本データセットは、ATOMIC 常識データセットで定義されたテンプレート文と関係を収めたものである。分割は 3 種類(train、test、dev)ある。
データは次の URL からダウンロードできる: https://maartensap.com/atomic/data/atomic_data.tgz
収録ファイルは以下のとおり:
v4_atomic_all_agg.csv: イベントを 1 行に 1 件ずつ収め、すべてのアノテーションを 1 つのリストに集約したファイル(重複は除去されていないため、同じ値が重なって現れる場合がある)。v4_atomic_all.csv: どの作業者がどのアノテーションを付けたかを追跡する。各行は 1 人の作業者の回答のみであるため、同一イベントについて複数の行が存在する。v4_atomic_trn.csv,v4_atomic_dev.csv,v4_atomic_tst.csv: 上と同じ内容だが、train/dev/test の分割に従って振り分けたもの。
すべてのファイルは、以下の列を持つ CSV である:
- event: イベントの文字列表現そのもの。
- oEffect,oReact,oWant,xAttr,xEffect,xIntent,xNeed,xReact,xWant: 各次元のアノテーションであり、文字列の配列を JSON 形式で書き出したリストとして格納されている。
注記:
[""none""]は作業者が空の応答を明示的に返したことを示し、[]はその次元に作業者がアノテーションを付けなかったことを示す。 - prefix: 内容語の接頭辞列を表す JSON 形式のリスト(trn/dev/tst の分割を改善するために用いられた)。
- split: イベントが所属する分割を示す文字列表現。
データを pandas のデータフレームに読み込む際の推奨コード:
import pandas as pd
import json
df = pd.read_csv("v4_atomic_all.csv",index_col=0)
df.iloc[:,:9] = df.iloc[:,:9].apply(lambda col: col.apply(json.loads))
免責事項・内容に関する注意: ATOMIC のイベントは、さまざまな時期に書かれたブログ・物語・書籍から自動的に抽出されたものである。 イベントには暴力的または問題のある行動が描かれている可能性があるが、それらは(おそらく否定的ではあるものの依然として重要な)イベントに付随する常識的含意を学習させるために、コーパスに残してある。 明らかに時代遅れのイベントを少数取り除いたが、見落としがある可能性があるため、懸念がある場合は msap@cs.washington.edu に連絡されたい。
対応タスクとリーダーボード
[More Information Needed]
言語
en
データセットの構造
データの例
ATOMIC データセットからの例を 1 つ示す:
{'event': "PersonX uses PersonX's ___ to obtain", 'oEffect': [], 'oReact': ['annoyed', 'angry', 'worried'], 'oWant': [], 'prefix': ['uses', 'obtain'], 'split': 'trn', 'xAttr': [], 'xEffect': [], 'xIntent': ['to have an advantage', 'to fulfill a desire', 'to get out of trouble'], 'xNeed': [], 'xReact': ['pleased', 'smug', 'excited'], 'xWant': []}
データのフィールド
作者による注記:
- event: イベントを文字列として表現したもの。
- oEffect,oReact,oWant,xAttr,xEffect,xIntent,xNeed,xReact,xWant: 各次元のアノテーション。JSON 形式でダンプされた文字列として格納される。 注: "none" は作業者が空の応答を明示的に返したことを、[] は作業者がその次元にアノテーションしなかったことを表す。
- prefix: 内容語の接頭辞を表す、JSON 形式でダンプされた文字列(trn/dev/tst の分割を改善するために用いられた)。
- split: イベントが属する分割を示す文字列。
データの分割
ATOMIC データセットには、test、train、dev の 3 つの分割があり、形式は次のとおり:
データセットの作成
収集の根拠
本データセットは、常識推論を支援するために収集・作成された。
元データ
初期のデータ収集と正規化
詳細は研究論文とウェブサイトを参照されたい。本データセットは、ワシントン大学がクラウドソーシングで集めたデータを用いて作成された。
元の言語の生成者は誰か?
ATOMIC の著者とクラウドソーシングの作業者。
アノテーション
アノテーションの過程
フォームに沿って人手で行われたアノテーション。
アノテーターは誰か?
人手によるアノテーション。
個人情報と機微な情報
不明だが、おそらくなし。
データの利用に関する考慮事項
データセットの社会的影響
この研究の目的は、機械に常識を理解させることである。
バイアスについての議論
データは人手によるアノテーションなので、偏りが生じている可能性が高い。作者の注記は以下のとおり:
免責事項・内容に関する注意: ATOMIC のイベントは、さまざまな時期に書かれたブログ・物語・書籍から自動的に抽出されたものである。イベントには暴力的または問題のある行動が描かれている可能性があるが、それらは(おそらく否定的ではあるものの依然として重要な)イベントに付随する常識的含意を学習させるために、コーパスに残してある。明らかに時代遅れのイベントを少数取り除いたが、見落としがある可能性があるため、懸念がある場合は msap@cs.washington.edu に連絡されたい。
その他の既知の制限
関係の種類は多いが、データは非常に疎である。また、データセットの各項目は、oEffect、oRect などのさまざまな次元に沿って複数の文へ展開できる。
たとえば、イベント "PersonX uses PersonX's ___ to obtain" と次元 oReact の "annoyed" が与えられた場合、次のようなエントリに変換できる:
"PersonX uses PersonX's ___ to obtain => PersonY is annoyed"
追加情報
データセットのキュレーター
ワシントン大学の ATOMIC の著者たち。
ライセンス情報
ライセンスは Creative Commons Attribution 4.0 International License である。 https://creativecommons.org/licenses/by/4.0/
引用情報
@article{Sap2019ATOMICAA, title={ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning}, author={Maarten Sap and Ronan Le Bras and Emily Allaway and Chandra Bhagavatula and Nicholas Lourie and Hannah Rashkin and Brendan Roof and Noah A. Smith and Yejin Choi}, journal={ArXiv}, year={2019}, volume={abs/1811.00146} }
貢献者
このデータセットを追加してくれた @ontocord に感謝する。
原文(英語)
Dataset Card for An Atlas of Machine Commonsense for If-Then Reasoning - Atomic Common Sense Dataset
Dataset Summary
This dataset provides the template sentences and relationships defined in the ATOMIC common sense dataset. There are three splits - train, test, and dev.
Data can be downloaded here: https://maartensap.com/atomic/data/atomic_data.tgz
Files present:
v4_atomic_all_agg.csv: contains one event per line, with all annotations aggregated into one list (but not de-duplicated, so there might be repeats).v4_atomic_all.csv: keeps track of which worker did which annotations. Each line is the answers from one worker only, so there are multiple lines for the same event.v4_atomic_trn.csv,v4_atomic_dev.csv,v4_atomic_tst.csv: same as above, but split based on train/dev/test split.
All files are CSVs containing the following columns:
- event: just a string representation of the event.
- oEffect,oReact,oWant,xAttr,xEffect,xIntent,xNeed,xReact,xWant: annotations for each of the dimensions, stored in a json-dumped list of strings.
Note:
[""none""]means the worker explicitly responded with the empty response, whereas[]means the worker did not annotate this dimension. - prefix: json-dumped list that represents the prefix of content words (used to make a better trn/dev/tst split).
- split: string rep of which split the event belongs to.
Suggested code for loading the data into a pandas dataframe:
import pandas as pd
import json
df = pd.read_csv("v4_atomic_all.csv",index_col=0)
df.iloc[:,:9] = df.iloc[:,:9].apply(lambda col: col.apply(json.loads))
Disclaimer/Content warning: the events in atomic have been automatically extracted from blogs, stories and books written at various times. The events might depict violent or problematic actions, which we left in the corpus for the sake of learning the (probably negative but still important) commonsense implications associated with the events. We removed a small set of truly out-dated events, but might have missed some so please email us (msap@cs.washington.edu) if you have any concerns.
Supported Tasks and Leaderboards
[More Information Needed]
Languages
en
Dataset Structure
Data Instances
Here is one example from the atomic dataset:
{'event': "PersonX uses PersonX's ___ to obtain", 'oEffect': [], 'oReact': ['annoyed', 'angry', 'worried'], 'oWant': [], 'prefix': ['uses', 'obtain'], 'split': 'trn', 'xAttr': [], 'xEffect': [], 'xIntent': ['to have an advantage', 'to fulfill a desire', 'to get out of trouble'], 'xNeed': [], 'xReact': ['pleased', 'smug', 'excited'], 'xWant': []}
Data Fields
Notes from the authors:
- event: just a string representation of the event.
- oEffect,oReact,oWant,xAttr,xEffect,xIntent,xNeed,xReact,xWant: annotations for each of the dimensions, stored in a json-dumped string. Note: "none" means the worker explicitly responded with the empty response, whereas [] means the worker did not annotate this dimension.
- prefix: json-dumped string that represents the prefix of content words (used to make a better trn/dev/tst split).
- split: string rep of which split the event belongs to.
Data Splits
The atomic dataset has three splits: test, train and dev of the form:
Dataset Creation
Curation Rationale
This dataset was gathered and created over to assist in common sense reasoning.
Source Data
Initial Data Collection and Normalization
See the reaserch paper and website for more detail. The dataset was created by the University of Washington using crowd sourced data
Who are the source language producers?
The Atomic authors and crowd source.
Annotations
Annotation process
Human annotations directed by forms.
Who are the annotators?
Human annotations.
Personal and Sensitive Information
Unkown, but likely none.
Considerations for Using the Data
Social Impact of Dataset
The goal for the work is to help machines understand common sense.
Discussion of Biases
Since the data is human annotators, there is likely to be baised. From the authors:
Disclaimer/Content warning: the events in atomic have been automatically extracted from blogs, stories and books written at various times. The events might depict violent or problematic actions, which we left in the corpus for the sake of learning the (probably negative but still important) commonsense implications associated with the events. We removed a small set of truly out-dated events, but might have missed some so please email us (msap@cs.washington.edu) if you have any concerns.
Other Known Limitations
While there are many relationships, the data is quite sparse. Also, each item of the dataset could be expanded into multiple sentences along the vsrious dimensions, oEffect, oRect, etc.
For example, given event: "PersonX uses PersonX's ___ to obtain" and dimension oReact: "annoyed", this could be transformed into an entry:
"PersonX uses PersonX's ___ to obtain => PersonY is annoyed"
Additional Information
Dataset Curators
The authors of Aotmic at The University of Washington
Licensing Information
The Creative Commons Attribution 4.0 International License. https://creativecommons.org/licenses/by/4.0/
Citation Information
@article{Sap2019ATOMICAA, title={ATOMIC: An Atlas of Machine Commonsense for If-Then Reasoning}, author={Maarten Sap and Ronan Le Bras and Emily Allaway and Chandra Bhagavatula and Nicholas Lourie and Hannah Rashkin and Brendan Roof and Noah A. Smith and Yejin Choi}, journal={ArXiv}, year={2019}, volume={abs/1811.00146} }
Contributions
Thanks to @ontocord for adding this dataset.
- Downloads last month
- 83