Dataset Viewer

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.

allenai/peer_read(複製)

データの内容

科学論文の査読のデータ。ACL、NIPS、ICLR などの会議に投稿された 14,000 件を超える論文原稿と、その採択・不採択の判定を収めている。一部の論文には、専門家が書いた 10,000 件を超える査読のテキストが付いている。構成は論文の本文を解析した parsed_pdfs と、査読を並べた reviews の 2 つである。

データの概要

項目 値
行数 24,728(2 構成・6 分割の合計)
列数 2〜12(構成による)
容量 32.02 KB(ファイル 4 本)
ファイルの形式 json 1、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
parsed_pdfs train 11,090、test 637、validation 637 2
reviews train 11,090、test 637、validation 637 12

列(構成 parsed_pdfs。構成ごとに列が違う)

列 型
name string
metadata struct{source, title, authors, emails, sections}

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/peer_read をリビジョン 53e19322c88bb01c1b0c6a61bde68bc2b1c3028e 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

論文

アノテーションの方法

元のカードにアノテーションの節はあるが、中身は空欄(More Information Needed)だった。

微調整に使うときの注意

  • 調査メモの台帳での区分: D(記載なし・other・unknown・資料間の食い違い)。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
science_tech.md Q4. 査読スコア・採否(score / noul) (L119) license タグ unknown(HF API)。GitHub は「ライセンス上の制約で一部のセクションは取得手順のみ」(README 確認) 確認済 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: peer_read.py のみ(HF API)。ACL / NIPS / ICLR / arXiv の混合。2017 年以前で古い

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。


peer_read 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

データセットの概要

PeerRead は、研究者がこの重要な成果物を研究できるよう公開された、科学論文の査読データセットである。ACL、NIPS、ICLR などのトップティア会議に投稿された 14,000 件を超える論文原稿と、それに対応する採択・不採択の判定を含む。さらに、その一部の論文については、専門家が書いた 10,000 件を超えるテキスト形式の査読を収めている。

対応タスクとリーダーボード

[More Information Needed]

言語

en-English

データセットの構造

データの例

[More Information Needed]

データのフィールド

parsed_pdfs

  • name: string データセット内のファイル名
  • metadata: dict 論文のメタデータ
    • source: string 論文の出典
    • authors: list<string> 論文の著者の一覧
    • title: string 論文の題名
    • sections: list<dict> セクションの見出しと、それに対応する説明の一覧
      • heading: string セクションの見出し
      • text: string セクションの説明
    • references: string 参考文献の一覧
      • title: string 参照先の論文の題名
      • author: list<string> 参照先の論文の著者の一覧
      • venue: string 参照先の掲載媒体
      • citeRegEx: string 参照先の citeRegEx
      • shortCiteRegEx: string 参照先の shortCiteRegEx
      • year: int 参照先の論文の発表年
    • referenceMentions: list<string> 参照の言及の一覧
      • referenceID: int 参照の言及の ID
      • context: string 参照の言及の文脈
      • startOffset: int 参照の開始オフセット
      • endOffset: int 参照の終了オフセット
    • year: int 論文の発表年
    • abstractText: string 論文の要旨
    • creator: string 論文の作成者

reviews

  • id: int 査読の ID
  • conference: string 会議名
  • comments: string 査読コメント
  • subjects: string 査読の対象分野
  • version: string 査読のバージョン
  • date_of_submission: string 投稿日
  • title: string 論文の題名
  • authors: list<string> 論文の著者の一覧
  • accepted: bool 論文が採択されたかを示すフラグ
  • abstract: string 論文の要旨
  • histories: list<string> リンク付きの論文の詳細情報
  • reviews: dict 論文に対する査読
    • date: string 査読の日付
    • title: string 論文の題名
    • other_keys: string 査読者のその他の詳細
    • originality: string 独創性の評点
    • comments: string 査読者のコメント
    • is_meta_review: bool 査読の種類を示すフラグ
    • recommendation: string 査読者の推奨
    • replicability: string 再現性の評点
    • presentation_format: string 発表形式
    • clarity: string 明瞭さの評点
    • meaningful_comparison: string 意味のある比較の評点
    • substance: string 内容の充実度の評点
    • reviewer_confidence: string 査読者の確信度の評点
    • soundness_correctness: string 健全性・正しさの評点
    • appropriateness: string 適切性の評点
    • impact: string インパクトの評点

データの分割

[More Information Needed]

データセットの作成

収集の根拠

[More Information Needed]

元データ

初期のデータ収集と正規化

[More Information Needed]

元の言語の生成者は誰か?

[More Information Needed]

アノテーション

アノテーションの過程

[More Information Needed]

アノテーターは誰か?

[More Information Needed]

個人情報と機微な情報

[More Information Needed]

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

[More Information Needed]

バイアスについての議論

[More Information Needed]

その他の既知の制限

[More Information Needed]

追加情報

データセットのキュレーター

Dongyeop Kang, Waleed Ammar, Bhavana Dalvi Mishra, Madeleine van Zuylen, Sebastian Kohlmeier, Eduard Hovy, Roy Schwartz

ライセンス情報

[More Information Needed]

引用情報

@inproceedings{kang18naacl, title = {A Dataset of Peer Reviews (PeerRead): Collection, Insights and NLP Applications}, author = {Dongyeop Kang and Waleed Ammar and Bhavana Dalvi and Madeleine van Zuylen and Sebastian Kohlmeier and Eduard Hovy and Roy Schwartz}, booktitle = {Meeting of the North American Chapter of the Association for Computational Linguistics (NAACL)}, address = {New Orleans, USA}, month = {June}, url = {https://arxiv.org/abs/1804.09635}, year = {2018} }

貢献者

このデータセットを追加してくれた @vinaykudari に感謝する。

原文(英語)

Dataset Card for peer_read

Dataset Summary

PearRead is a dataset of scientific peer reviews available to help researchers study this important artifact. The dataset consists of over 14K paper drafts and the corresponding accept/reject decisions in top-tier venues including ACL, NIPS and ICLR, as well as over 10K textual peer reviews written by experts for a subset of the papers.

Supported Tasks and Leaderboards

[More Information Needed]

Languages

en-English

Dataset Structure

Data Instances

[More Information Needed]

Data Fields

parsed_pdfs

  • name: string Filename in the dataset
  • metadata: dict Paper metadata
    • source: string Paper source
    • authors: list<string> List of paper authors
    • title: string Paper title
    • sections: list<dict> List of section heading and corresponding description
      • heading: string Section heading
      • text: string Section description
    • references: string List of references
      • title: string Title of reference paper
      • author: list<string> List of reference paper authors
      • venue: string Reference venue
      • citeRegEx: string Reference citeRegEx
      • shortCiteRegEx: string Reference shortCiteRegEx
      • year: int Reference publish year
    • referenceMentions: list<string> List of reference mentions
      • referenceID: int Reference mention ID
      • context: string Reference mention context
      • startOffset: int Reference startOffset
      • endOffset: int Reference endOffset
    • year: int Paper publish year
    • abstractText: string Paper abstract
    • creator: string Paper creator

reviews

  • id: int Review ID
  • conference: string Conference name
  • comments: string Review comments
  • subjects: string Review subjects
  • version: string Review version
  • date_of_submission: string Submission date
  • title: string Paper title
  • authors: list<string> List of paper authors
  • accepted: bool Paper accepted flag
  • abstract: string Paper abstract
  • histories: list<string> Paper details with link
  • reviews: dict Paper reviews
    • date: string Date of review
    • title: string Paper title
    • other_keys: string Reviewer other details
    • originality: string Originality score
    • comments: string Reviewer comments
    • is_meta_review: bool Review type flag
    • recommendation: string Reviewer recommendation
    • replicability: string Replicability score
    • presentation_format: string Presentation type
    • clarity: string Clarity score
    • meaningful_comparison: string Meaningful comparison score
    • substance: string Substance score
    • reviewer_confidence: string Reviewer confidence score
    • soundness_correctness: string Soundness correctness score
    • appropriateness: string Appropriateness score
    • impact: string Impact score

Data Splits

[More Information Needed]

Dataset Creation

Curation Rationale

[More Information Needed]

Source Data

Initial Data Collection and Normalization

[More Information Needed]

Who are the source language producers?

[More Information Needed]

Annotations

Annotation process

[More Information Needed]

Who are the annotators?

[More Information Needed]

Personal and Sensitive Information

[More Information Needed]

Considerations for Using the Data

Social Impact of Dataset

[More Information Needed]

Discussion of Biases

[More Information Needed]

Other Known Limitations

[More Information Needed]

Additional Information

Dataset Curators

Dongyeop Kang, Waleed Ammar, Bhavana Dalvi Mishra, Madeleine van Zuylen, Sebastian Kohlmeier, Eduard Hovy, Roy Schwartz

Licensing Information

[More Information Needed]

Citation Information

@inproceedings{kang18naacl, title = {A Dataset of Peer Reviews (PeerRead): Collection, Insights and NLP Applications}, author = {Dongyeop Kang and Waleed Ammar and Bhavana Dalvi and Madeleine van Zuylen and Sebastian Kohlmeier and Eduard Hovy and Roy Schwartz}, booktitle = {Meeting of the North American Chapter of the Association for Computational Linguistics (NAACL)}, address = {New Orleans, USA}, month = {June}, url = {https://arxiv.org/abs/1804.09635}, year = {2018} }

Contributions

Thanks to @vinaykudari for adding this dataset.

Downloads last month
47

Paper for HYUKI5/allenai__peer_read