Finetuned Cross Encoder L6 V2

🚀 クロスエンコーダ cross-encoder/ms-marco-MiniLM-L6-v2 をベースにしたもの

これは cross-encoder/ms-marco-MiniLM-L6-v2 を微調整した クロスエンコーダ モデルで、sentence-transformers ライブラリを使用しています。このモデルはテキストペアのスコアを計算でき、テキストの再ランキングや意味検索に利用できます。

🚀 クイックスタート

このモデルはテキストペアのスコアを計算するために使用でき、テキストの再ランキングや意味検索などのタスクに適しています。以下では、このモデルの使用方法を説明します。

✨ 主な機能

• cross-encoder/ms-marco-MiniLM-L6-v2 を微調整したもので、テキストペアのスコアを計算できます。
• テキストの再ランキングや意味検索タスクに適しています。

📦 インストール

まず、sentence-transformers ライブラリをインストールする必要があります。

pip install -U sentence-transformers

💻 使用例

基本的な使用法

from sentence_transformers import CrossEncoder

# Hugging Face Hub からモデルをダウンロード
model = CrossEncoder("CharlesPing/finetuned-cross-encoder-l6-v2")
# テキストペアのスコアを取得
pairs = [
    ['‘Getting hung up on the exact nature of the records is interesting, and there’s lots of technical work that can be done there, but the main take-home response there is that the trends we’ve been seeing since the 1970s are continuing and have not paused in any way,’ he said.”', 'Rosenzweig also criticized the "waffling—encouraged by the NPOV policy—[which] means that it is hard to discern any overall interpretive stance in Wikipedia history".'],
    ['After the 9/11 terrorist attacks grounded commercial air traffic, "there was a temperature drop while the airplanes weren\'t flying, for the week afterwards."', 'Play media At 9:42\xa0a.m., the Federal Aviation Administration (FAA) grounded all civilian aircraft within the continental U.S., and civilian aircraft already in flight were told to land immediately.'],
    ['But the central message of the IPCC AR4, is confirmed by the peer reviewed literature.', 'Scientific consensus is normally achieved through communication at conferences, publication in the scientific literature, replication (reproducible results by others), and peer review.'],
    ['"Many people think the science of climate change is settled.', 'During his administration, the bridge from Filadelfia and Liberia was constructed, as was the Old National Theater.'],
    ['“Even if you could calculate some sort of meaningful global temperature statistic, the figure would be unimportant.', 'Quantitative information or data is based on quantities obtained using a quantifiable measurement process.'],
]
scores = model.predict(pairs)
print(scores.shape)
# (5,)

# または、単一のテキストとの類似度に基づいて異なるテキストを並べ替える
ranks = model.rank(
    '‘Getting hung up on the exact nature of the records is interesting, and there’s lots of technical work that can be done there, but the main take-home response there is that the trends we’ve been seeing since the 1970s are continuing and have not paused in any way,’ he said.”',
    [
        'Rosenzweig also criticized the "waffling—encouraged by the NPOV policy—[which] means that it is hard to discern any overall interpretive stance in Wikipedia history".',
        'Play media At 9:42\xa0a.m., the Federal Aviation Administration (FAA) grounded all civilian aircraft within the continental U.S., and civilian aircraft already in flight were told to land immediately.',
        'Scientific consensus is normally achieved through communication at conferences, publication in the scientific literature, replication (reproducible results by others), and peer review.',
        'During his administration, the bridge from Filadelfia and Liberia was constructed, as was the Old National Theater.',
        'Quantitative information or data is based on quantities obtained using a quantifiable measurement process.',
    ]
)
# [{'corpus_id': ..., 'score': ...}, {'corpus_id': ..., 'score': ...}, ...]

📚 ドキュメント

モデル詳細

モデルの説明

モデルの出所

• ドキュメント：Sentence Transformers ドキュメント
• ドキュメント：クロスエンコーダドキュメント
• リポジトリ：GitHub 上の Sentence Transformers
• Hugging Face：Hugging Face 上のクロスエンコーダ

評価

指標

クロスエンコーダ再ランキング

• データセット：cross-rerank-dev-mixed-neg
• CrossEncoderRerankingEvaluator を使用して評価し、以下のパラメータを使用します。

  {
      "at_k": 10
  }
  

トレーニング詳細

トレーニングデータセット

未命名データセット

• サイズ：22,258 個のトレーニングサンプル
• 列：sentence_0、sentence_1、label
• 最初の 1000 個のサンプルに基づく近似統計情報： | | sentence_0 | sentence_1 | label | |------|------|------|------| | タイプ | 文字列 | 文字列 | 浮動小数点数 | | 詳細 |
  • 最小長：26 文字
  • 平均長：121.91 文字
  • 最大長：319 文字
  |
  • 最小長：36 文字
  • 平均長：140.85 文字
  • 最大長：573 文字
  |
  • 最小値：0.0
  • 平均値：0.16
  • 最大値：1.0
  |
• サンプル： | sentence_0 | sentence_1 | label | |------|------|------| | ‘Getting hung up on the exact nature of the records is interesting, and there’s lots of technical work that can be done there, but the main take-home response there is that the trends we’ve been seeing since the 1970s are continuing and have not paused in any way,’ he said.” | Rosenzweig also criticized the "waffling—encouraged by the NPOV policy—[which] means that it is hard to discern any overall interpretive stance in Wikipedia history". | 1.0 | | After the 9/11 terrorist attacks grounded commercial air traffic, "there was a temperature drop while the airplanes weren't flying, for the week afterwards." | Play media At 9:42 a.m., the Federal Aviation Administration (FAA) grounded all civilian aircraft within the continental U.S., and civilian aircraft already in flight were told to land immediately. | 1.0 | | But the central message of the IPCC AR4, is confirmed by the peer reviewed literature. | Scientific consensus is normally achieved through communication at conferences, publication in the scientific literature, replication (reproducible results by others), and peer review. | 1.0 |
• 損失関数：FitMixinLoss

トレーニングハイパーパラメータ

非デフォルトのハイパーパラメータ

• eval_strategy：ステップごとに評価
• per_device_train_batch_size：16
• per_device_eval_batch_size：16

すべてのハイパーパラメータ

トレーニングログ

フレームワークバージョン

• Python: 3.11.12
• Sentence Transformers: 4.1.0
• Transformers: 4.51.3
• PyTorch: 2.6.0+cu124
• Accelerate: 1.6.0
• Datasets: 3.5.1
• Tokenizers: 0.21.1

🔧 技術詳細

このモデルは cross-encoder/ms-marco-MiniLM-L6-v2 を微調整し、sentence-transformers ライブラリを使用してトレーニングされています。トレーニング過程では特定のハイパーパラメータと損失関数を使用し、テキスト再ランキングタスクでのモデル性能を最適化しています。

📖 引用

BibTeX

Sentence Transformers

@inproceedings{reimers-2019-sentence-bert,
    title = "Sentence-BERT: Sentence Embeddings using Siamese BERT-Networks",
    author = "Reimers, Nils and Gurevych, Iryna",
    booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing",
    month = "11",
    year = "2019",
    publisher = "Association for Computational Linguistics",
    url = "https://arxiv.org/abs/1908.10084",
}