{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/codex-a-comprehensive-knowledge-graph","title":"CoDEx: A Comprehensive Knowledge Graph Completion Benchmark","arxiv_id":"2009.07810","date":"2020-09-16","proceeding":"EMNLP 2020 11","authors":["Tara Safavi","Danai Koutra"],"abstract":"We present CoDEx, a set of knowledge graph completion datasets extracted from Wikidata and Wikipedia that improve upon existing knowledge graph completion benchmarks in scope and level of difficulty. In terms of scope, CoDEx comprises three knowledge graphs varying in size and structure, multilingual descriptions of entities and relations, and tens of thousands of hard negative triples that are plausible but verified to be false. To characterize CoDEx, we contribute thorough empirical analyses and benchmarking experiments. First, we analyze each CoDEx dataset in terms of logical relation patterns. Next, we report baseline link prediction and triple classification results on CoDEx for five extensively tuned embedding models. Finally, we differentiate CoDEx from the popular FB15K-237 knowledge graph completion dataset by showing that CoDEx covers more diverse and interpretable content, and is a more difficult link prediction benchmark. Data, code, and pretrained models are available at https://bit.ly/2EPbrJs.","url_abs":"https://arxiv.org/abs/2009.07810v2","url_pdf":"https://arxiv.org/pdf/2009.07810v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"codex-a-comprehensive-knowledge-graph","repo_url":"https://github.com/tsafavi/codex","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"codex-a-comprehensive-knowledge-graph","repo_url":"https://github.com/facebookresearch/ssl-relation-prediction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}}],"tasks":[{"task_slug":"benchmarking","task_name":"Benchmarking"},{"task_slug":"knowledge-graph-completion","task_name":"Knowledge Graph Completion"},{"task_slug":"knowledge-graphs","task_name":"Knowledge Graphs"},{"task_slug":"link-prediction","task_name":"Link Prediction"},{"task_slug":"triple-classification","task_name":"Triple Classification"}],"methods":[],"datasets_introduced":[{"slug":"codex-large","name":"CoDEx Large","full_name":""},{"slug":"codex-medium","name":"CoDEx Medium","full_name":""},{"slug":"codex","name":"CoDEx Small","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/link-prediction-on-codex-large","task":"Link Prediction","dataset":"CoDEx Large","model":"TuckER","rank_in_archive_order":2,"of":6,"metrics":{"Hits@1":"0.244","Hits@10":"0.430","Hits@3":"0.3395","MRR":"0.309"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex-large","task":"Link Prediction","dataset":"CoDEx Large","model":"RESCAL","rank_in_archive_order":3,"of":6,"metrics":{"Hits@1":"0.242","Hits@10":"0.419","Hits@3":"0.3313","MRR":"0.304"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex-large","task":"Link Prediction","dataset":"CoDEx Large","model":"ConvE","rank_in_archive_order":4,"of":6,"metrics":{"Hits@1":"0.240","Hits@10":"0.420","Hits@3":"0.3298","MRR":"0.303"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex-large","task":"Link Prediction","dataset":"CoDEx Large","model":"ComplEx","rank_in_archive_order":5,"of":6,"metrics":{"Hits@1":"0.237","Hits@10":"0.400","Hits@3":"0.3179","MRR":"0.294"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex-large","task":"Link Prediction","dataset":"CoDEx Large","model":"TransE","rank_in_archive_order":6,"of":6,"metrics":{"Hits@1":"0.116","Hits@10":"0.317","Hits@3":"0.2188","MRR":"0.187"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex-medium","task":"Link Prediction","dataset":"CoDEx Medium","model":"ComplEx","rank_in_archive_order":3,"of":7,"metrics":{"Hits@1":"0.244","Hits@10":"0.456","Hits@3":"0.3477","MRR":"0.337"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex-medium","task":"Link Prediction","dataset":"CoDEx Medium","model":"TuckER","rank_in_archive_order":4,"of":7,"metrics":{"Hits@1":"0.223","Hits@10":"0.454","Hits@3":"0.3363","MRR":"0.328"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex-medium","task":"Link Prediction","dataset":"CoDEx Medium","model":"ConvE","rank_in_archive_order":5,"of":7,"metrics":{"Hits@1":"0.262","Hits@10":"0.476","Hits@3":"0.3701","MRR":"0.318"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex-medium","task":"Link Prediction","dataset":"CoDEx Medium","model":"RESCAL","rank_in_archive_order":6,"of":7,"metrics":{"Hits@1":"0.239","Hits@10":"0.464","Hits@3":"0.3551","MRR":"0.317"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex-medium","task":"Link Prediction","dataset":"CoDEx Medium","model":"TransE","rank_in_archive_order":7,"of":7,"metrics":{"Hits@1":"0.259","Hits@10":"0.458","Hits@3":"0.3599","MRR":"0.303"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex","task":"Link Prediction","dataset":"CoDEx Small","model":"TuckER","rank_in_archive_order":2,"of":6,"metrics":{"Hits@1":"0.372","Hits@10":"0.646","Hits@3":"0.5038","MRR":"0.444"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex","task":"Link Prediction","dataset":"CoDEx Small","model":"ConvE","rank_in_archive_order":3,"of":6,"metrics":{"Hits@1":"0.219","Hits@10":"0.634","Hits@3":"0.4218","MRR":"0.444"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex","task":"Link Prediction","dataset":"CoDEx Small","model":"RESCAL","rank_in_archive_order":4,"of":6,"metrics":{"Hits@1":"0.343","Hits@10":"0.635","Hits@3":"0.4926","MRR":"0.404"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex","task":"Link Prediction","dataset":"CoDEx Small","model":"ComplEx","rank_in_archive_order":5,"of":6,"metrics":{"Hits@1":"0.293","Hits@10":"0.623","Hits@3":"0.4494","MRR":"0.404"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-codex","task":"Link Prediction","dataset":"CoDEx Small","model":"TransE","rank_in_archive_order":6,"of":6,"metrics":{"Hits@1":"0.339","Hits@10":"0.638","Hits@3":"0.4975","MRR":"0.354"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2009.07810","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}