fix issues in the dataset loading script

SciDuet.py

@@ -81,7 +81,8 @@ class SciDuet(datasets.GeneratorBasedBuilder):
                     "slide_id": datasets.Value("string"),
                     "slide_title": datasets.Value("string"),
                     "slide_content_text": datasets.Value("string"),
-
+                    "target": datasets.Value("string"),
+                    "references": [datasets.Value("string")],
                 }
             ),
             supervised_keys=None,
@@ -151,7 +152,7 @@ class SciDuet(datasets.GeneratorBasedBuilder):
                     slide_content_text = '\n'.join(item["slides"][j]["text"])
 
                     yield id_, {
-                        "gem_id": gem_id,
+                        "gem_id": id_,
                         "paper_id": paper_id,
                         "paper_title": paper_title,
                         "paper_abstract": paper_abstract,
@@ -160,6 +161,8 @@ class SciDuet(datasets.GeneratorBasedBuilder):
                         "slide_id": id_,
                         "slide_title": slide_title,
                         "slide_content_text": slide_content_text,
+                        "target": slide_content_text,
+                        "references": [] if split == "train" else [slide_content_text],
                 }
 
 

dataset_infos.json

@@ -0,0 +1 @@
+{"default": {"description": "SciDuet is the first publicaly available dataset for the challenging task of document2slides generation,\nThe dataset integrated into GEM is the ACL portion of the whole dataset described in \"https://aclanthology.org/2021.naacl-main.111.pdf\".\nIt contains the full Dev and Test sets, and a portion of the Train dataset. \nWe additionally create a challenge dataset in which the slide titles do not match with the \nsection headers of the corresponding paper.\nNote that although we cannot release the whole training dataset due to copyright issues, researchers can still \nuse our released data procurement code from https://github.com/IBM/document2slides\nto generate the training dataset from the online ICML/NeurIPS anthologies. \nIn the released dataset, the original papers and slides (both are in PDF format) are carefully processed by a combination of PDF/Image processing tookits.\nThe text contents from multiple slides that correspond to the same slide title are mreged. \n", "citation": "@inproceedings{sun-etal-2021-d2s,\n    title = \"{D}2{S}: Document-to-Slide Generation Via Query-Based Text Summarization\",\n    author = \"Sun, Edward  and\n      Hou, Yufang  and\n      Wang, Dakuo  and\n      Zhang, Yunfeng  and\n      Wang, Nancy X. R.\",\n    booktitle = \"Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies\",\n    month = June,\n    year = \"2021\",\n    address = \"Online\",\n    publisher = \"Association for Computational Linguistics\",\n    url = \"https://aclanthology.org/2021.naacl-main.111\",\n    doi = \"10.18653/v1/2021.naacl-main.111\",\n    pages = \"1405--1418\",\n}\n", "homepage": "", "license": "Apache License 2.0", "features": {"gem_id": {"dtype": "string", "id": null, "_type": "Value"}, "paper_id": {"dtype": "string", "id": null, "_type": "Value"}, "paper_title": {"dtype": "string", "id": null, "_type": "Value"}, "paper_abstract": {"dtype": "string", "id": null, "_type": "Value"}, "paper_content": {"feature": {"paper_content_id": {"dtype": "int32", "id": null, "_type": "Value"}, "paper_content_text": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "paper_headers": {"feature": {"paper_header_number": {"dtype": "string", "id": null, "_type": "Value"}, "paper_header_content": {"dtype": "string", "id": null, "_type": "Value"}}, "length": -1, "id": null, "_type": "Sequence"}, "slide_id": {"dtype": "string", "id": null, "_type": "Value"}, "slide_title": {"dtype": "string", "id": null, "_type": "Value"}, "slide_content_text": {"dtype": "string", "id": null, "_type": "Value"}, "target": {"dtype": "string", "id": null, "_type": "Value"}, "references": [{"dtype": "string", "id": null, "_type": "Value"}]}, "post_processed": null, "supervised_keys": null, "task_templates": null, "builder_name": "sci_duet", "config_name": "default", "version": {"version_str": "1.0.0", "description": null, "major": 1, "minor": 0, "patch": 0}, "splits": {"train": {"name": "train", "num_bytes": 52934618, "num_examples": 1869, "dataset_name": "sci_duet"}, "validation": {"name": "validation", "num_bytes": 20166909, "num_examples": 740, "dataset_name": "sci_duet"}, "test": {"name": "test", "num_bytes": 35037908, "num_examples": 1203, "dataset_name": "sci_duet"}, "challenge_woSectionHeader": {"name": "challenge_woSectionHeader", "num_bytes": 25889890, "num_examples": 895, "dataset_name": "sci_duet"}}, "download_checksums": {"train.json": {"num_bytes": 8518173, "checksum": "11f89a3d7bfcf1bf6094f23052e6c11fb26605e63d3b515d14cbad5421a96a7b"}, "validation.json": {"num_bytes": 3315786, "checksum": "7aff9a0667603f480842423b1f37b6ca8a06f278d9af9a099d7c978b1ff75b86"}, "test.json": {"num_bytes": 5174341, "checksum": "57d366dbfd1d0e9c272d780a45ed7e6ee00085bebadef567e32266c34114fc05"}, "challenge_woSectionHeader.json": {"num_bytes": 4956971, "checksum": "d8f87eb3ba861e21178601c51a9fc6bb48cf735467fdb976ebc37764687a1aea"}}, "download_size": 21965271, "post_processing_size": null, "dataset_size": 134029325, "size_in_bytes": 155994596}}