{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,12]],"date-time":"2026-03-12T13:05:11Z","timestamp":1773320711524,"version":"3.50.1"},"reference-count":47,"publisher":"MIT Press","license":[{"start":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T00:00:00Z","timestamp":1773187200000},"content-version":"vor","delay-in-days":6,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"content-domain":{"domain":["direct.mit.edu"],"crossmark-restriction":true},"short-container-title":[],"published-print":{"date-parts":[[2026,3,5]]},"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Stance detection on social media plays a vital role in understanding public opinion on contentious topics. While prior work leverages external knowledge sources like Wikipedia to enrich limited target information, it primarily introduces conceptual content, neglecting the interpretability potential of knowledge and often leading to the incorporation of irrelevant or redundant information that hinders stance prediction performance. To address this, we introduce PiKGL, a Pruned interpretable Knowledge Graph Learning framework for explainable stance detection. Specifically, we first extract event triplets and topics to obtain real-world knowledge, which is then used to construct an interpretable knowledge graph. To ensure precision and minimize noise, we introduce a retrieval-guided pruning strategy that incorporates commonsense knowledge, filtering redundant information of the interpretable knowledge graph. Finally, the pruned knowledge graph is injected into a large language model to jointly model textual, target, and commonsense for improved stance comprehension. Experimental results conducted on three public datasets demonstrate our PiKGL achieves state-of-the-art performance on stance detection.<\/jats:p>","DOI":"10.1162\/tacl.a.612","type":"journal-article","created":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T19:37:11Z","timestamp":1773257831000},"page":"217-232","update-policy":"https:\/\/doi.org\/10.1162\/mitpressjournals.corrections.policy","source":"Crossref","is-referenced-by-count":0,"title":["PiKGL: Leveraging Pruned Knowledge Graphs for Explainable Stance\n                    Detection"],"prefix":"10.1162","volume":"14","author":[{"given":"Bingbing","family":"Wang","sequence":"first","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"},{"name":"The Hong Kong Polytechnic University, Hong Kong, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jingjie","family":"Lin","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhixin","family":"Bai","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xintong","family":"Song","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qianlong","family":"Wang","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Min","family":"Yang","sequence":"additional","affiliation":[{"name":"SIAT, Chinese Academy of Sciences, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xi","family":"Zeng","sequence":"additional","affiliation":[{"name":"The 30th Research Institute of China Electronics Technology Group Corporation, Chengdu, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jing","family":"Li","sequence":"additional","affiliation":[{"name":"The Hong Kong Polytechnic University, Hong Kong, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ruifeng","family":"Xu","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Shenzhen, China"},{"name":"Peng Cheng Laboratory, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","published-online":{"date-parts":[[2026,3,5]]},"reference":[{"key":"2026031115370742200_bib1","doi-asserted-by":"publisher","first-page":"8913","DOI":"10.18653\/v1\/2020.emnlp-main.717","article-title":"Zero-shot stance detection: A dataset and\n                        model using generalized topic representations","volume-title":"Proceedings of the 2020 Conference on Empirical Methods in Natural\n                        Language Processing (EMNLP)","author":"Allaway","year":"2020"},{"key":"2026031115370742200_bib2","doi-asserted-by":"publisher","first-page":"902","DOI":"10.1162\/tacl_a_00580","article-title":"Reasoning over public and private data in\n                        retrieval-based systems","volume":"11","author":"Arora","year":"2023","journal-title":"Transactions of the\n                        Association for Computational Linguistics"},{"key":"2026031115370742200_bib3","doi-asserted-by":"publisher","first-page":"876","DOI":"10.18653\/v1\/D16-1084","article-title":"Stance detection with bidirectional\n                        conditional encoding","volume-title":"Proceedings of the 2016\n                        Conference on Empirical Methods in Natural Language Processing","author":"Augenstein","year":"2016"},{"key":"2026031115370742200_bib4","doi-asserted-by":"publisher","first-page":"3615","DOI":"10.18653\/v1\/D19-1371","article-title":"Scibert: A pretrained language model for\n                        scientific text","volume-title":"Proceedings of the 2019\n                        Conference on Empirical Methods in Natural Language Processing and the 9th\n                        International Joint Conference on Natural Language Processing\n                        (EMNLP-IJCNLP)","author":"Iz","year":"2019"},{"key":"2026031115370742200_bib5","doi-asserted-by":"publisher","first-page":"1","DOI":"10.1109\/IJCNN60899.2024.10651178","article-title":"Sentkb-bert: Sentiment-filtered\n                        knowledge-based stance detection","volume-title":"2024\n                        International Joint Conference on Neural Networks (IJCNN)","author":"Chen","year":"2024"},{"key":"2026031115370742200_bib6","article-title":"Bge m3-embedding: Multi-lingual, multi-functionality,\n                        multi-granularity text embeddings through self-knowledge\n                        distillation","author":"Chen","year":"2024","journal-title":"arXiv preprint\n                        arXiv:2402.03216"},{"key":"2026031115370742200_bib7","doi-asserted-by":"publisher","first-page":"177","DOI":"10.18653\/v1\/2024.knowllm-1.15","article-title":"Retrieval-augmented generation in\n                        multilingual settings","volume-title":"Proceedings of the 1st\n                        Workshop on Towards Knowledgeable Language Models (KnowLLM 2024)","author":"Chirkova","year":"2024"},{"key":"2026031115370742200_bib8","doi-asserted-by":"publisher","first-page":"824","DOI":"10.1007\/978-981-99-6207-5_26","article-title":"Adversarial network with external\n                        knowledge for zero-shot stance detection","volume-title":"Proceedings of the 22nd Chinese National Conference on Computational\n                        Linguistics","author":"Chunling","year":"2023"},{"key":"2026031115370742200_bib9","doi-asserted-by":"publisher","first-page":"283","DOI":"10.1162\/tacl_a_00644","article-title":"Evaluating the ripple effects of knowledge editing in\n                        language models","volume":"12","author":"Cohen","year":"2024","journal-title":"Transactions of the Association for\n                        Computational Linguistics"},{"key":"2026031115370742200_bib10","doi-asserted-by":"publisher","first-page":"4171","DOI":"10.18653\/v1\/N19-1423","article-title":"BERT: Pre-training of deep bidirectional\n                        transformers for language understanding","volume-title":"Proceedings of the 2019 Conference of the North American Chapter of\n                        the Association for Computational Linguistics: Human Language Technologies,\n                        Volume 1 (Long and Short Papers)","author":"Devlin","year":"2019"},{"key":"2026031115370742200_bib11","article-title":"The llama 3 herd of models","author":"Dubey","year":"2024","journal-title":"arXiv preprint arXiv:2407.21783"},{"key":"2026031115370742200_bib12","article-title":"Bertopic: Neural topic modeling with a\n                        class-based tf-idf procedure","author":"Grootendorst","year":"2022","journal-title":"arXiv preprint\n                        arXiv:2203.05794"},{"key":"2026031115370742200_bib13","doi-asserted-by":"publisher","first-page":"2020","DOI":"10.1145\/3404835.3463001","article-title":"Rumor detection on social media with event\n                        augmentations","volume-title":"Proceedings of the 44th\n                        International ACM SIGIR Conference on Research and Development in\n                        Information Retrieval","author":"He","year":"2021"},{"key":"2026031115370742200_bib14","doi-asserted-by":"publisher","first-page":"71","DOI":"10.18653\/v1\/2022.wassa-1.7","article-title":"Infusing knowledge from wikipedia to\n                        enhance stance detection","volume-title":"Proceedings of the 12th\n                        Workshop on Computational Approaches to Subjectivity, Sentiment &\n                        Social Media Analysis","author":"He","year":"2022"},{"key":"2026031115370742200_bib15","article-title":"Mistral 7b","author":"Jiang","year":"2023","journal-title":"arXiv\n                        preprint arXiv:2310.06825"},{"key":"2026031115370742200_bib16","doi-asserted-by":"publisher","first-page":"837","DOI":"10.1145\/3477495.3531979","article-title":"Few-shot stance detection via target-aware\n                        prompt distillation","volume-title":"Proceedings of the 45th\n                        International ACM SIGIR Conference on Research and Development in\n                        Information Retrieval","author":"Jiang","year":"2022"},{"key":"2026031115370742200_bib17","doi-asserted-by":"publisher","first-page":"6769","DOI":"10.18653\/v1\/2020.emnlp-main.550","article-title":"Dense passage retrieval for open-domain question\n                        answering","volume-title":"2020 Conference on Empirical Methods\n                        in Natural Language Processing, EMNLP 2020","author":"Karpukhin","year":"2020"},{"key":"2026031115370742200_bib18","doi-asserted-by":"publisher","first-page":"1392","DOI":"10.1162\/tacl_a_00525","article-title":"Scientia potentia est\u2013on the role\n                        of knowledge in computational argumentation","volume":"10","author":"Lauscher","year":"2022","journal-title":"Transactions of the Association for Computational\n                        Linguistics"},{"key":"2026031115370742200_bib19","first-page":"1","article-title":"Gdelt: Global data on events, location,\n                        and tone, 1979\u20132012","volume-title":"ISA Annual\n                        Convention","author":"Leetaru","year":"2013"},{"key":"2026031115370742200_bib20","doi-asserted-by":"publisher","first-page":"15703","DOI":"10.18653\/v1\/2023.emnlp-main.972","article-title":"Stance detection on social media with\n                        background knowledge","volume-title":"Proceedings of the 2023\n                        Conference on Empirical Methods in Natural Language Processing","author":"Li","year":"2023"},{"key":"2026031115370742200_bib21","doi-asserted-by":"publisher","first-page":"2355","DOI":"10.18653\/v1\/2021.findings-acl.208","article-title":"P-stance: A large dataset for stance\n                        detection in political domain","volume-title":"Findings of the\n                        Association for Computational Linguistics: ACL-IJCNLP 2021","author":"Li","year":"2021"},{"key":"2026031115370742200_bib22","doi-asserted-by":"publisher","first-page":"3152","DOI":"10.18653\/v1\/2021.findings-acl.278","article-title":"Enhancing zero-shot and few-shot stance detection with\n                        commonsense knowledge graph","volume-title":"Findings of the\n                        Association for Computational Linguistics: ACL-IJCNLP 2021","author":"Liu","year":"2021"},{"key":"2026031115370742200_bib23","doi-asserted-by":"publisher","first-page":"2901","DOI":"10.1609\/aaai.v34i03.5681","article-title":"K-bert: Enabling language representation with knowledge\n                        graph","volume-title":"Proceedings of the AAAI Conference on\n                        Artificial Intelligence","author":"Liu","year":"2020"},{"key":"2026031115370742200_bib24","article-title":"Roberta: A robustly optimized bert pretraining\n                        approach","volume":"364","author":"Liu","year":"2019","journal-title":"arXiv preprint arXiv:1907.11692"},{"key":"2026031115370742200_bib25","first-page":"7112","article-title":"Exploiting sentiment and common sense for\n                        zero-shot stance detection","volume-title":"Proceedings of the\n                        29th International Conference on Computational Linguistics","author":"Luo","year":"2022"},{"key":"2026031115370742200_bib26","doi-asserted-by":"publisher","first-page":"9","DOI":"10.18653\/v1\/2020.emnlp-demos.2","article-title":"Bertweet: A pre-trained language model for\n                        English tweets","volume-title":"Proceedings of the 2020\n                        Conference on Empirical Methods in Natural Language Processing: System\n                        Demonstrations","author":"Nguyen","year":"2020"},{"key":"2026031115370742200_bib27","doi-asserted-by":"publisher","first-page":"12921","DOI":"10.18653\/v1\/2023.findings-emnlp.861","article-title":"COMET-M: Reasoning about multiple events\n                        in complex sentences","volume-title":"Findings of the Association\n                        for Computational Linguistics: EMNLP 2023","author":"Ravi","year":"2023"},{"issue":"10","key":"2026031115370742200_bib28","doi-asserted-by":"publisher","first-page":"11229","DOI":"10.1609\/aaai.v36i10.21373","article-title":"Cem: Commonsense-aware empathetic response\n                        generation","volume":"36","author":"Sabour","year":"2022","journal-title":"Proceedings of the AAAI Conference on\n                        Artificial Intelligence"},{"key":"2026031115370742200_bib29","doi-asserted-by":"publisher","first-page":"4615","DOI":"10.18653\/v1\/2020.emnlp-main.373","article-title":"Unsupervised commonsense question answering with\n                        self-talk","volume-title":"Proceedings of the 2020 Conference on\n                        Empirical Methods in Natural Language Processing (EMNLP)","author":"Shwartz","year":"2020"},{"key":"2026031115370742200_bib30","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v31i1.11164","article-title":"Conceptnet 5.5: An open multilingual graph\n                        of general knowledge","volume-title":"Proceedings of the AAAI\n                        Conference on Artificial Intelligence","author":"Speer","year":"2017"},{"key":"2026031115370742200_bib31","doi-asserted-by":"publisher","first-page":"1173","DOI":"10.1145\/3331184.3331367","article-title":"Modeling transferable topics for cross-target stance\n                        detection","volume-title":"Proceedings of the 42nd International\n                        ACM SIGIR Conference on Research and Development in Information\n                        Retrieval","author":"Wei","year":"2019"},{"key":"2026031115370742200_bib32","doi-asserted-by":"publisher","first-page":"1491","DOI":"10.18653\/v1\/2023.acl-short.127","article-title":"Zero-shot and few-shot stance detection on\n                        varied topics via conditional generation","volume-title":"Proceedings of the 61st Annual Meeting of the Association for\n                        Computational Linguistics (Volume 2: Short Papers)","author":"Wen","year":"2023"},{"key":"2026031115370742200_bib33","doi-asserted-by":"publisher","first-page":"778","DOI":"10.18653\/v1\/P18-2123","article-title":"Cross-target stance classification with\n                        self-attention networks","volume-title":"Proceedings of the 56th\n                        Annual Meeting of the Association for Computational Linguistics (Volume 2:\n                        Short Papers)","author":"Chang","year":"2018"},{"issue":"3","key":"2026031115370742200_bib34","doi-asserted-by":"publisher","first-page":"682","DOI":"10.26599\/BDMA.2024.9020021","article-title":"Collaborative knowledge infusion for\n                        low-resource stance detection","volume":"7","author":"Yan","year":"2024","journal-title":"Big Data Mining and\n                        Analytics"},{"key":"2026031115370742200_bib35","article-title":"Qwen2. 5 technical report","author":"An","year":"2024","journal-title":"arXiv preprint arXiv:2412.15115"},{"key":"2026031115370742200_bib36","article-title":"Xlnet: Generalized autoregressive pretraining for language\n                        understanding","volume-title":"Advances in Neural Information\n                        Processing Systems","author":"Yang","year":"2019"},{"key":"2026031115370742200_bib37","doi-asserted-by":"publisher","first-page":"131","DOI":"10.1007\/978-3-031-17120-8_11","article-title":"Kformer: Knowledge injection in transformer\n                        feed-forward layers","volume-title":"CCF International Conference\n                        on Natural Language Processing and Chinese Computing","author":"Yao","year":"2022"},{"key":"2026031115370742200_bib38","doi-asserted-by":"publisher","first-page":"11630","DOI":"10.1609\/aaai.v36i10.21417","article-title":"Jaket: Joint pre-training of knowledge graph and language\n                        understanding","volume-title":"Proceedings of the AAAI Conference\n                        on Artificial Intelligence","author":"Donghan","year":"2022"},{"key":"2026031115370742200_bib39","article-title":"How would stance detection techniques evolve after the launch\n                        of chatgpt?","author":"Zhang","year":"2022","journal-title":"arXiv preprint arXiv:2212.14548"},{"key":"2026031115370742200_bib40","doi-asserted-by":"publisher","first-page":"9820","DOI":"10.18653\/v1\/2024.emnlp-main.548","article-title":"Extract, define, canonicalize: An LLM-based framework for\n                        knowledge graph construction","volume-title":"Proceedings of the\n                        2024 Conference on Empirical Methods in Natural Language\n                    Processing","author":"Zhang","year":"2024"},{"key":"2026031115370742200_bib41","doi-asserted-by":"publisher","first-page":"126943","DOI":"10.1016\/j.neucom.2023.126943","article-title":"Commonsense-based adversarial learning framework for\n                        zero-shot stance detection","volume":"563","author":"Zhang","year":"2024","journal-title":"Neurocomputing"},{"key":"2026031115370742200_bib42","doi-asserted-by":"publisher","first-page":"21456","DOI":"10.18653\/v1\/2024.emnlp-main.1196","article-title":"Synthetic knowledge ingestion: Towards\n                        knowledge refinement and injection for enhancing large language\n                        models","volume-title":"Proceedings of the 2024 Conference on\n                        Empirical Methods in Natural Language Processing","author":"Zhang","year":"2024"},{"key":"2026031115370742200_bib43","doi-asserted-by":"publisher","first-page":"1441","DOI":"10.18653\/v1\/P19-1139","article-title":"Ernie: Enhanced language representation with informative\n                        entities","volume-title":"Proceedings of the 57th Annual Meeting\n                        of the Association for Computational Linguistics","author":"Zhang","year":"2019"},{"key":"2026031115370742200_bib44","doi-asserted-by":"publisher","first-page":"371","DOI":"10.18653\/v1\/2024.naacl-short.32","article-title":"Llm-driven knowledge injection advances\n                        zero-shot and cross-target stance detection","volume-title":"Proceedings of the 2024 Conference of the North American Chapter of\n                        the Association for Computational Linguistics: Human Language Technologies\n                        (Volume 2: Short Papers)","author":"Zhang","year":"2024"},{"key":"2026031115370742200_bib45","doi-asserted-by":"publisher","first-page":"15697","DOI":"10.18653\/v1\/2024.acl-long.838","article-title":"EZ-STANCE: A large dataset for English\n                        zero-shot stance detection","volume-title":"Proceedings of the\n                        62nd Annual Meeting of the Association for Computational Linguistics (Volume\n                        1: Long Papers)","author":"Zhao","year":"2024"},{"key":"2026031115370742200_bib46","doi-asserted-by":"publisher","first-page":"13369","DOI":"10.18653\/v1\/2023.acl-long.747","article-title":"C-STANCE: A large dataset for Chinese\n                        zero-shot stance detection","volume-title":"Proceedings of the\n                        61st Annual Meeting of the Association for Computational Linguistics (Volume\n                        1: Long Papers)","author":"Zhao","year":"2023"},{"key":"2026031115370742200_bib47","doi-asserted-by":"publisher","first-page":"2070","DOI":"10.1145\/3477495.3531807","article-title":"Enhancing zero-shot stance detection via\n                        targeted background knowledge","volume-title":"Proceedings of the\n                        45th International ACM SIGIR Conference on Research and Development in\n                        Information Retrieval","author":"Zhu","year":"2022"}],"container-title":["Transactions of the Association for Computational Linguistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/direct.mit.edu\/tacl\/article-pdf\/doi\/10.1162\/TACL.a.612\/2587252\/tacl.a.612.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"syndication"},{"URL":"https:\/\/direct.mit.edu\/tacl\/article-pdf\/doi\/10.1162\/TACL.a.612\/2587252\/tacl.a.612.pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T19:37:13Z","timestamp":1773257833000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/tacl\/article\/doi\/10.1162\/TACL.a.612\/135721\/PiKGL-Leveraging-Pruned-Knowledge-Graphs-for"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,5]]},"references-count":47,"URL":"https:\/\/doi.org\/10.1162\/tacl.a.612","relation":{},"ISSN":["2307-387X"],"issn-type":[{"value":"2307-387X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,5]]}}}