| {"Blaise-g--PubMed_summ": {"description": "\n PubMed dataset for summarization.\n From paper: A Discourse-Aware Attention Model for Abstractive Summarization of Long Documents\" by A. Cohan et al.\n See: https://aclanthology.org/N18-2097.pdf \n See: https://github.com/armancohan/long-summarization\n", "citation": " @inproceedings{cohan-etal-2018-discourse,\n title = \"A Discourse-Aware Attention Model for Abstractive Summarization of Long Documents\",\n author = \"Cohan, Arman and\n Dernoncourt, Franck and\n Kim, Doo Soon and\n Bui, Trung and\n Kim, Seokhwan and\n Chang, Walter and\n Goharian, Nazli\",\n booktitle = \"Proceedings of the 2018 Conference of the North {A}merican Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 2 (Short Papers)\",\n month = jun,\n year = \"2018\",\n address = \"New Orleans, Louisiana\",\n publisher = \"Association for Computational Linguistics\",\n url = \"https://aclanthology.org/N18-2097\",\n doi = \"10.18653/v1/N18-2097\",\n pages = \"615--621\",\n abstract = \"Neural abstractive summarization models have led to promising results in summarizing relatively short documents. We propose the first model for abstractive summarization of single, longer-form documents (e.g., research papers). Our approach consists of a new hierarchical encoder that models the discourse structure of a document, and an attentive discourse-aware decoder to generate the summary. Empirical results on two large-scale datasets of scientific papers show that our model significantly outperforms state-of-the-art models.\",\n}\n", "homepage": "https://github.com/armancohan/long-summarization", "license": "", "features": {"article": {"dtype": "string", "id": null, "_type": "Value"}, "abstract": {"dtype": "string", "id": null, "_type": "Value"}}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "pubmed-summarization", "config_name": "section", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 2155856847.1349354, "num_examples": 119874, "dataset_name": "PubMed_summ"}, "test": {"name": "test", "num_bytes": 121825977.20411535, "num_examples": 6655, "dataset_name": "PubMed_summ"}, "validation": {"name": "validation", "num_bytes": 121890595.73586613, "num_examples": 6628, "dataset_name": "PubMed_summ"}}, "download_checksums": null, "download_size": 1118917236, "post_processing_size": null, "dataset_size": 2399573420.074917, "size_in_bytes": 3518490656.074917}} |