finosfoundation/finreg_esma_json_and_code_benchmark_synthetic_data

该数据集包含多个配置，每个配置具有不同的特征。主要特征包括文档ID、文档文本、文档文件名、文档元数据（如文件大小）、原始摘要、摘要、文档摘要、摘要模型、文本块信息、多跳文本块信息等。具体配置如下：

chunked: 包含文档分割成的块及其摘要。
finreg_code: 包含文档、代码块和JSON块信息。
finreg_json: 包含文档、JSON块信息。
ingested: 包含文档信息。
multi_hop_questions: 包含多跳问题的相关信息。
single_shot_questions: 包含单次问答的信息。
summarized: 包含文档摘要信息。

The dataset consists of multiple configurations, each with different features. Main features include document ID, document text, document filename, document metadata (such as file size), raw summaries, summaries, document summaries, summarization models, text chunk information, multi-hop text chunk information, etc. Specific configurations are as follows:

chunked: Contains chunks of documents and their summaries.
finreg_code: Contains documents, code blocks, and JSON blocks information.
finreg_json: Contains documents and JSON blocks information.
ingested: Contains document information.
multi_hop_questions: Contains information related to multi-hop questions.
single_shot_questions: Contains information related to single-shot questions.
summarized: Contains document summary information.