allenai/multinews_sparse_max
收藏资源简介:
--- annotations_creators: - expert-generated language_creators: - expert-generated language: - en license: - other multilinguality: - monolingual pretty_name: Multi-News size_categories: - 10K<n<100K source_datasets: - original task_categories: - summarization task_ids: - news-articles-summarization paperswithcode_id: multi-news train-eval-index: - config: default task: summarization task_id: summarization splits: train_split: train eval_split: test col_mapping: document: text summary: target metrics: - type: rouge name: Rouge --- This is a copy of the [Multi-News](https://huggingface.co/datasets/multi_news) dataset, except the input source documents of its `test` split have been replaced by a __sparse__ retriever. The retrieval pipeline used: - __query__: The `summary` field of each example - __corpus__: The union of all documents in the `train`, `validation` and `test` splits - __retriever__: BM25 via [PyTerrier](https://pyterrier.readthedocs.io/en/latest/) with default settings - __top-k strategy__: `"max"`, i.e. the number of documents retrieved, `k`, is set as the maximum number of documents seen across examples in this dataset, in this case `k==10` Retrieval results on the `train` set: | Recall@100 | Rprec | Precision@k | Recall@k | | ----------- | ----------- | ----------- | ----------- | | 0.8793 | 0.7460 | 0.2213 | 0.8264 | Retrieval results on the `validation` set: | Recall@100 | Rprec | Precision@k | Recall@k | | ----------- | ----------- | ----------- | ----------- | | 0.8748 | 0.7453 | 0.2173 | 0.8232 | Retrieval results on the `test` set: | Recall@100 | Rprec | Precision@k | Recall@k | | ----------- | ----------- | ----------- | ----------- | | 0.8775 | 0.7480 | 0.2187 | 0.8250 |
annotations_creators: - 专家生成 language_creators: - 专家生成 language: - 英语(en) license: - 其他 multilinguality: - 单语 pretty_name: Multi-News size_categories: - 10000 < 样本数量 < 100000 source_datasets: - 原生数据集 task_categories: - 文本摘要 task_ids: - 新闻文章摘要 paperswithcode_id: multi-news train-eval-index: - config: 默认(default) task: 文本摘要 task_id: summarization splits: train_split: train eval_split: test col_mapping: document: 文本 summary: 目标 metrics: - type: ROUGE name: Rouge 本数据集为[Multi-News](https://huggingface.co/datasets/multi_news)数据集的复刻版本,仅其测试划分(test split)的输入源文档已被替换为稀疏检索器(sparse retriever)的检索结果。所采用的检索流程如下: - 查询(query):每个样本的`summary`字段 - 检索语料库(corpus):训练、验证与测试划分中所有文档的并集 - 检索器(retriever):基于[PyTerrier](https://pyterrier.readthedocs.io/en/latest/)实现的BM25算法,采用默认参数配置 - Top-K选取策略(top-k strategy):采用"max"模式,即检索文档数`k`被设置为该数据集中所有样本所需的最大文档数,本次配置中`k=10` 训练集检索效果: | 召回率@100 | R精度(Rprec) | 精确率@k | 召回率@k | | ----------- | ----------- | ----------- | ----------- | | 0.8793 | 0.7460 | 0.2213 | 0.8264 | 验证集检索效果: | 召回率@100 | R精度(Rprec) | 精确率@k | 召回率@k | | ----------- | ----------- | ----------- | ----------- | | 0.8748 | 0.7453 | 0.2173 | 0.8232 | 测试集检索效果: | 召回率@100 | R精度(Rprec) | 精确率@k | 召回率@k | | ----------- | ----------- | ----------- | ----------- | | 0.8775 | 0.7480 | 0.2187 | 0.8250 |
数据集概述
- 名称: Multi-News
- 语言: 英语 (
en) - 许可证: 其他 (
other) - 多语言性: 单语 (
monolingual) - 大小: 10K<n<100K
- 来源: 原始 (
original) - 任务类别: 摘要生成 (
summarization) - 任务ID: news-articles-summarization
- 训练与评估配置:
- 任务: 摘要生成
- 任务ID: summarization
- 分割:
- 训练分割:
train - 评估分割:
test
- 训练分割:
- 列映射:
- 文档:
text - 摘要:
target
- 文档:
- 评估指标:
- 类型:
rouge - 名称:
Rouge
- 类型:
数据集特点
- 注释创建者: 专家生成 (
expert-generated) - 语言创建者: 专家生成 (
expert-generated)



