{"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/mocosa-momentum-contrast-for-knowledge-graph","title":"MoCoSA: Momentum Contrast for Knowledge Graph Completion with Structure-Augmented Pre-trained Language Models","arxiv_id":"2308.08204","date":"2023-08-16","proceeding":null,"authors":["Jiabang He","Liu Jia","Lei Wang","Xiyao Li","Xing Xu"],"abstract":"Knowledge Graph Completion (KGC) aims to conduct reasoning on the facts within knowledge graphs and automatically infer missing links. Existing methods can mainly be categorized into structure-based or description-based. On the one hand, structure-based methods effectively represent relational facts in knowledge graphs using entity embeddings. However, they struggle with semantically rich real-world entities due to limited structural information and fail to generalize to unseen entities. On the other hand, description-based methods leverage pre-trained language models (PLMs) to understand textual information. They exhibit strong robustness towards unseen entities. However, they have difficulty with larger negative sampling and often lag behind structure-based methods. To address these issues, in this paper, we propose Momentum Contrast for knowledge graph completion with Structure-Augmented pre-trained language models (MoCoSA), which allows the PLM to perceive the structural information by the adaptable structure encoder. To improve learning efficiency, we proposed momentum hard negative and intra-relation negative sampling. Experimental results demonstrate that our approach achieves state-of-the-art performance in terms of mean reciprocal rank (MRR), with improvements of 2.5% on WN18RR and 21% on OpenBG500.","url_abs":"https://arxiv.org/abs/2308.08204v1","url_pdf":"https://arxiv.org/pdf/2308.08204v1.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":[],"tasks":[{"task_slug":"entity-embeddings","task_name":"Entity Embeddings"},{"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"}],"methods":[{"method_slug":"fail","method_name":"fail"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/link-prediction-on-fb15k-237","task":"Link Prediction","dataset":"FB15k-237","model":"MoCoSA","rank_in_archive_order":8,"of":75,"metrics":{"Hits@1":"0.292","Hits@10":"0.578","Hits@3":"0.42","MRR":"0.387"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-openbg500","task":"Link Prediction","dataset":"OpenBG500","model":"MoCoSA","rank_in_archive_order":1,"of":1,"metrics":{"Hits@1":"0.531","Hits@10":"0.83","Hits@3":"0.711","MRR":"0.634"},"uses_additional_data":false},{"leaderboard":"/sota/link-prediction-on-wn18rr","task":"Link Prediction","dataset":"WN18RR","model":"MoCoSA","rank_in_archive_order":3,"of":75,"metrics":{"Hits@1":"0.624","Hits@10":"0.82","Hits@3":"0.737","MRR":"0.696"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2308.08204","atlas_url":"https://app.syntology.ai/?focus=2308.08204","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}