multixscience_dense_max

This is a copy of the [Multi-XScience](https://huggingface.co/datasets/multi_x_science_sum) dataset, except the input source documents of its `test` split have been replaced by a __dense__ retriever. The retrieval pipeline used:

- __query__: The `related_work` field of each example
- __corpus__: The union of all documents in the `train`, `validation` and `test` splits
- __retriever__: [`facebook/contriever-msmarco`](https://huggingface.co/facebook/contriever-msmarco) 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==20`

Retrieval results on the `train` set:

| Recall@100 | Rprec | Precision@k | Recall@k |
| ----------- | ----------- | ----------- | ----------- |
| 0.5270 | 0.2005 | 0.0573 | 0.3785 |

Retrieval results on the `validation` set:

| Recall@100 | Rprec | Precision@k | Recall@k |
| ----------- | ----------- | ----------- | ----------- |
| 0.5310 | 0.2026 | 0.059 | 0.3831 |

Retrieval results on the `test` set:

| Recall@100 | Rprec | Precision@k | Recall@k |
| ----------- | ----------- | ----------- | ----------- |
| 0.5229 | 0.2081 | 0.058 | 0.3794 |

本数据集为[Multi-XScience](https://huggingface.co/datasets/multi_x_science_sum)数据集的副本，仅将其`test`划分的输入源文档替换为**稠密**检索器获取的内容。所用检索流水线如下：

- **查询（query）**：每个样本的`related_work`字段
- **语料库（corpus）**：`train`、`validation`与`test`划分下所有文档的并集
- **检索器（retriever）**：基于[PyTerrier](https://pyterrier.readthedocs.io/en/latest/)框架，采用默认配置加载[`facebook/contriever-msmarco`](https://huggingface.co/facebook/contriever-msmarco)模型
- **top-k策略**：采用`"max"`策略，即检索文档数`k`被设置为该数据集所有样本中所需文档的最大数量，本案例中`k=20`

训练集检索结果：
| 召回率@100 | R精度 | 精确率@k | 召回率@k |
| ----------- | ----------- | ----------- | ----------- |
| 0.5270 | 0.2005 | 0.0573 | 0.3785 |

验证集检索结果：
| 召回率@100 | R精度 | 精确率@k | 召回率@k |
| ----------- | ----------- | ----------- | ----------- |
| 0.5310 | 0.2026 | 0.059 | 0.3831 |

测试集检索结果：
| 召回率@100 | R精度 | 精确率@k | 召回率@k |
| ----------- | ----------- | ----------- | ----------- |
| 0.5229 | 0.2081 | 0.058 | 0.3794 |