{"url":"/task/knowledge-base-question-answering","name":"Knowledge Base Question Answering","slug":"knowledge-base-question-answering","description_markdown":"Knowledge Base Q&A is the task of answering questions from a knowledge base.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Modeling Semantics with Gated Graph Neural Networks for Knowledge Base Question Answering](https://www.aclweb.org/anthology/C18-1280.pdf) )</span>","categories":[{"name":"Miscellaneous","url":"/area/miscellaneous"},{"name":"Natural Language Processing","url":"/area/natural-language-processing"},{"name":"Reasoning","url":"/area/reasoning"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":136,"papers_with_code":67,"benchmarks":10,"benchmark_tables_in_archive":10,"benchmark_tables_shown":10,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":15,"subtasks":0,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/knowledge-base-question-answering-on-qald-9","slug":"knowledge-base-question-answering-on-qald-9","dataset":"QALD-9-Plus","dataset_url":"/dataset/qald-9-plus","rows_in_archive":12,"metrics":["Macro F1"],"first_row_in_archive_order":{"model":"QAnswer-Wikidata-English","paper_title":"QALD-9-plus: A Multilingual Dataset for Question Answering over DBpedia and Wikidata Translated by Native Speakers","paper_url":"/paper/qald-9-plus-a-multilingual-dataset-for","paper_date":"2022-01-31","arxiv_id":"2202.00120","code_links":[{"title":"perevalov/qald_9_plus","url":"https://github.com/perevalov/qald_9_plus"}],"syntology":null}},{"leaderboard":"/sota/knowledge-base-question-answering-on-1","slug":"knowledge-base-question-answering-on-1","dataset":"WebQuestionsSP","dataset_url":"/dataset/webquestionssp","rows_in_archive":8,"metrics":["Hits@1","F1","Accuracy"],"first_row_in_archive_order":{"model":"ChatKBQA","paper_title":"ChatKBQA: A Generate-then-Retrieve Framework for Knowledge Base Question Answering with Fine-tuned Large Language Models","paper_url":"/paper/chatkbqa-a-generate-then-retrieve-framework","paper_date":"2023-10-13","arxiv_id":"2310.08975","code_links":[{"title":"lhrlab/chatkbqa","url":"https://github.com/lhrlab/chatkbqa"}],"syntology":{"n":16,"n_ran":8,"n_unverified":8,"n_pointer_only":0}}},{"leaderboard":"/sota/knowledge-base-question-answering-on-lc-quad","slug":"knowledge-base-question-answering-on-lc-quad","dataset":"LC-QuAD 1.0","dataset_url":"/dataset/lc-quad-2-0","rows_in_archive":7,"metrics":["F1"],"first_row_in_archive_order":{"model":"HGNet (oracle entity used)","paper_title":"Outlining and Filling: Hierarchical Query Graph Generation for Answering Complex Questions over Knowledge Graphs","paper_url":"/paper/outlining-and-filling-hierarchical-query","paper_date":"2021-11-01","arxiv_id":"2111.00732","code_links":[{"title":"bahuia/aqgnet","url":"https://github.com/bahuia/aqgnet"},{"title":"bahuia/hgnet","url":"https://github.com/bahuia/hgnet"}],"syntology":null}},{"leaderboard":"/sota/knowledge-base-question-answering-on","slug":"knowledge-base-question-answering-on","dataset":"ComplexWebQuestions","dataset_url":"/dataset/complexwebquestions","rows_in_archive":6,"metrics":["Accuracy","F1","Hits@1","EM"],"first_row_in_archive_order":{"model":"ChatKBQA","paper_title":"ChatKBQA: A Generate-then-Retrieve Framework for Knowledge Base Question Answering with Fine-tuned Large Language Models","paper_url":"/paper/chatkbqa-a-generate-then-retrieve-framework","paper_date":"2023-10-13","arxiv_id":"2310.08975","code_links":[{"title":"lhrlab/chatkbqa","url":"https://github.com/lhrlab/chatkbqa"}],"syntology":{"n":16,"n_ran":8,"n_unverified":8,"n_pointer_only":0}}},{"leaderboard":"/sota/knowledge-base-question-answering-on-4","slug":"knowledge-base-question-answering-on-4","dataset":"SimpleQuestionsWikiData","dataset_url":"/dataset/simplequestionswikidata","rows_in_archive":5,"metrics":["F1"],"first_row_in_archive_order":{"model":"SPARKLE","paper_title":"SPARKLE: Enhancing SPARQL Generation with Direct KG Integration in Decoding","paper_url":"/paper/sparkle-enhancing-sparql-generation-with","paper_date":"2024-06-29","arxiv_id":"2407.01626","code_links":[{"title":"zzaebok/sparkle","url":"https://github.com/zzaebok/sparkle"}],"syntology":null}},{"leaderboard":"/sota/knowledge-base-question-answering-on-grailqa","slug":"knowledge-base-question-answering-on-grailqa","dataset":"GrailQA","dataset_url":"/dataset/grailqa","rows_in_archive":2,"metrics":["Compositional EM","Compositional F1","EM","I.I.D. EM","I.I.D. F1","Overall EM","Overall F1","Zero-shot EM","Zero-shot F1"],"first_row_in_archive_order":{"model":"ReTraCk","paper_title":"ReTraCk: A Flexible and Efficient Framework for Knowledge Base Question Answering","paper_url":"/paper/retrack-a-flexible-and-efficient-framework","paper_date":"2021-08-01","arxiv_id":null,"code_links":[{"title":"microsoft/KC","url":"https://github.com/microsoft/KC/tree/main/papers/ReTraCk"}],"syntology":null}},{"leaderboard":"/sota/knowledge-base-question-answering-on-2","slug":"knowledge-base-question-answering-on-2","dataset":"SimpleQuestions","dataset_url":"/dataset/simplequestions","rows_in_archive":1,"metrics":["EM"],"first_row_in_archive_order":{"model":"PoG-GPT4 (Tan et al., 2024)","paper_title":"Paths-over-Graph: Knowledge Graph Empowered Large Language Model Reasoning","paper_url":"/paper/paths-over-graph-knowledge-graph-enpowered","paper_date":"2024-10-18","arxiv_id":"2410.14211","code_links":[{"title":"SteveTANTAN/PoG","url":"https://github.com/SteveTANTAN/PoG"}],"syntology":{"n":12,"n_ran":1,"n_unverified":11,"n_pointer_only":0}}},{"leaderboard":"/sota/knowledge-base-question-answering-on-3","slug":"knowledge-base-question-answering-on-3","dataset":"WebQuestions","dataset_url":"/dataset/webquestions","rows_in_archive":1,"metrics":["EM"],"first_row_in_archive_order":{"model":"PoG-GPT4 (Tan et al., 2024)","paper_title":"Paths-over-Graph: Knowledge Graph Empowered Large Language Model Reasoning","paper_url":"/paper/paths-over-graph-knowledge-graph-enpowered","paper_date":"2024-10-18","arxiv_id":"2410.14211","code_links":[{"title":"SteveTANTAN/PoG","url":"https://github.com/SteveTANTAN/PoG"}],"syntology":{"n":12,"n_ran":1,"n_unverified":11,"n_pointer_only":0}}},{"leaderboard":"/sota/knowledge-base-question-answering-on-webqsp","slug":"knowledge-base-question-answering-on-webqsp","dataset":"WebQSP-WD","dataset_url":"/dataset/webqsp","rows_in_archive":1,"metrics":["Avg F1"],"first_row_in_archive_order":{"model":"GGNN","paper_title":"Modeling Semantics with Gated Graph Neural Networks for Knowledge Base Question Answering","paper_url":"/paper/modeling-semantics-with-gated-graph-neural","paper_date":"2018-08-13","arxiv_id":"1808.04126","code_links":[{"title":"UKPLab/coling2018-graph-neural-networks-question-answering","url":"https://github.com/UKPLab/coling2018-graph-neural-networks-question-answering"}],"syntology":null}},{"leaderboard":"/sota/knowledge-base-question-answering-on-webqsp-1","slug":"knowledge-base-question-answering-on-webqsp-1","dataset":"WebQSP","dataset_url":"/dataset/webqsp","rows_in_archive":1,"metrics":["EM"],"first_row_in_archive_order":{"model":"PoG-GPT4 (Tan et al., 2024)","paper_title":"Paths-over-Graph: Knowledge Graph Empowered Large Language Model Reasoning","paper_url":"/paper/paths-over-graph-knowledge-graph-enpowered","paper_date":"2024-10-18","arxiv_id":"2410.14211","code_links":[{"title":"SteveTANTAN/PoG","url":"https://github.com/SteveTANTAN/PoG"}],"syntology":{"n":12,"n_ran":1,"n_unverified":11,"n_pointer_only":0}}}],"datasets":[{"url":"/dataset/webquestions","name":"WebQuestions","full_name":"","num_papers_in_archive":241},{"url":"/dataset/simplequestions","name":"SimpleQuestions","full_name":"","num_papers_in_archive":120},{"url":"/dataset/complexwebquestions","name":"ComplexWebQuestions","full_name":"ComplexWebQuestions","num_papers_in_archive":62},{"url":"/dataset/webquestionssp","name":"WebQuestionsSP","full_name":"WebQuestions Semantic Parses Dataset","num_papers_in_archive":61},{"url":"/dataset/grailqa","name":"GrailQA","full_name":"Strongly Generalizable Question Answering","num_papers_in_archive":34},{"url":"/dataset/lc-quad-2-0","name":"LC-QuAD","full_name":"Largescale Complex Question Answering Dataset","num_papers_in_archive":10},{"url":"/dataset/rubq","name":"RuBQ","full_name":"Russian Knowledge Base Questions","num_papers_in_archive":7},{"url":"/dataset/frenchmedmcqa","name":"FrenchMedMCQA","full_name":"FrenchMedMCQA: A French Multiple-Choice Question Answering Dataset for Medical domain","num_papers_in_archive":6},{"url":"/dataset/simplequestionswikidata","name":"SimpleQuestionsWikiData","full_name":"","num_papers_in_archive":6},{"url":"/dataset/paraqa","name":"ParaQA","full_name":"","num_papers_in_archive":5},{"url":"/dataset/webqsp","name":"WebQSP","full_name":null,"num_papers_in_archive":5},{"url":"/dataset/vanilla","name":"VANiLLa","full_name":"","num_papers_in_archive":4},{"url":"/dataset/qald-9-plus","name":"QALD-9-Plus","full_name":"","num_papers_in_archive":3},{"url":"/dataset/geoquestions1089","name":"GeoQuestions1089","full_name":"","num_papers_in_archive":1},{"url":"/dataset/orkg-qa","name":"ORKG-QA","full_name":"","num_papers_in_archive":1}],"subtasks":[],"parent_tasks":[{"url":"/task/question-answering","name":"Question Answering"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":30,"of":67,"tagged_in_all":136,"items":[{"url":"/paper/spinach-sparql-based-information-navigation","title":"SPINACH: SPARQL-Based Information Navigation for Challenging Real-World Questions","date":"2024-07-16","arxiv_id":"2407.11417","repositories_listed":2,"syntology":null},{"url":"/paper/evaluation-of-chatgpt-as-a-question-answering","title":"Can ChatGPT Replace Traditional KBQA Models? An In-depth Analysis of the Question Answering Performance of the GPT LLM Family","date":"2023-03-14","arxiv_id":"2303.07992","repositories_listed":2,"syntology":null},{"url":"/paper/don-t-generate-discriminate-a-proposal-for","title":"Don't Generate, Discriminate: A Proposal for Grounding Language Models to Real-World Environments","date":"2022-12-19","arxiv_id":"2212.09736","repositories_listed":2,"syntology":{"n":3,"n_ran":1,"n_unverified":2,"n_pointer_only":3}},{"url":"/paper/outlining-and-filling-hierarchical-query","title":"Outlining and Filling: Hierarchical Query Graph Generation for Answering Complex Questions over Knowledge Graphs","date":"2021-11-01","arxiv_id":"2111.00732","repositories_listed":2,"syntology":null},{"url":"/paper/kagnet-knowledge-aware-graph-networks-for","title":"KagNet: Knowledge-Aware Graph Networks for Commonsense Reasoning","date":"2019-09-04","arxiv_id":"1909.02151","repositories_listed":2,"syntology":{"n":12,"n_ran":1,"n_unverified":11,"n_pointer_only":0}},{"url":"/paper/bidirectional-attentive-memory-networks-for","title":"Bidirectional Attentive Memory Networks for Question Answering over Knowledge Bases","date":"2019-03-06","arxiv_id":"1903.02188","repositories_listed":2,"syntology":{"n":11,"n_ran":0,"n_unverified":11,"n_pointer_only":0}},{"url":"/paper/multi-task-learning-with-multi-view-attention","title":"Multi-Task Learning with Multi-View Attention for Answer Selection and Knowledge Base Question Answering","date":"2018-12-06","arxiv_id":"1812.02354","repositories_listed":2,"syntology":null},{"url":"/paper/beyond-seen-data-improving-kbqa","title":"Beyond Seen Data: Improving KBQA Generalization Through Schema-Guided Logical Form Generation","date":"2025-02-18","arxiv_id":"2502.12737","repositories_listed":1,"syntology":null},{"url":"/paper/kbqa-o1-agentic-knowledge-base-question","title":"KBQA-o1: Agentic Knowledge Base Question Answering with Monte Carlo Tree Search","date":"2025-01-31","arxiv_id":"2501.18922","repositories_listed":1,"syntology":{"n":8,"n_ran":2,"n_unverified":6,"n_pointer_only":0}},{"url":"/paper/cypherbench-towards-precise-retrieval-over","title":"CypherBench: Towards Precise Retrieval over Full-scale Modern Knowledge Graphs in the LLM Era","date":"2024-12-24","arxiv_id":"2412.18702","repositories_listed":1,"syntology":null},{"url":"/paper/paths-over-graph-knowledge-graph-enpowered","title":"Paths-over-Graph: Knowledge Graph Empowered Large Language Model Reasoning","date":"2024-10-18","arxiv_id":"2410.14211","repositories_listed":1,"syntology":{"n":12,"n_ran":1,"n_unverified":11,"n_pointer_only":0}},{"url":"/paper/semantic-parsing-with-candidate-expressions","title":"Semantic Parsing with Candidate Expressions for Knowledge Base Question Answering","date":"2024-10-01","arxiv_id":"2410.00414","repositories_listed":1,"syntology":null},{"url":"/paper/2408-02337","title":"Developing PUGG for Polish: A Modern Approach to KBQA, MRC, and IR Dataset Construction","date":"2024-08-05","arxiv_id":"2408.02337","repositories_listed":1,"syntology":null},{"url":"/paper/sparkle-enhancing-sparql-generation-with","title":"SPARKLE: Enhancing SPARQL Generation with Direct KG Integration in Decoding","date":"2024-06-29","arxiv_id":"2407.01626","repositories_listed":1,"syntology":null},{"url":"/paper/understand-what-llm-needs-dual-preference","title":"Understand What LLM Needs: Dual Preference Alignment for Retrieval-Augmented Generation","date":"2024-06-26","arxiv_id":"2406.18676","repositories_listed":1,"syntology":{"n":3,"n_ran":1,"n_unverified":2,"n_pointer_only":3}},{"url":"/paper/retinaqa-a-knowledge-base-question-answering","title":"RetinaQA: A Robust Knowledge Base Question Answering Model for both Answerable and Unanswerable Questions","date":"2024-03-16","arxiv_id":"2403.10849","repositories_listed":1,"syntology":null},{"url":"/paper/interactive-kbqa-multi-turn-interactions-for","title":"Interactive-KBQA: Multi-Turn Interactions for Knowledge Base Question Answering with Large Language Models","date":"2024-02-23","arxiv_id":"2402.15131","repositories_listed":1,"syntology":{"n":10,"n_ran":7,"n_unverified":3,"n_pointer_only":10}},{"url":"/paper/triad-a-framework-leveraging-a-multi-role-llm","title":"Triad: A Framework Leveraging a Multi-Role LLM-based Agent to Solve Knowledge Base Question Answering","date":"2024-02-22","arxiv_id":"2402.14320","repositories_listed":1,"syntology":{"n":5,"n_ran":5,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/clue-guided-path-exploration-an-efficient","title":"Fine-Grained Stateful Knowledge Exploration: A Novel Paradigm for Integrating Knowledge Graphs with Large Language Models","date":"2024-01-24","arxiv_id":"2401.13444","repositories_listed":1,"syntology":null},{"url":"/paper/relation-aware-question-answering-for","title":"Relation-Aware Question Answering for Heterogeneous Knowledge Graphs","date":"2023-12-19","arxiv_id":"2312.11922","repositories_listed":1,"syntology":null},{"url":"/paper/combining-transfer-learning-with-in-context","title":"Few-shot Transfer Learning for Knowledge Base Question Answering: Fusing Supervised Models with In-Context Learning","date":"2023-11-15","arxiv_id":"2311.08894","repositories_listed":1,"syntology":null},{"url":"/paper/benchmarking-geospatial-question-answering","title":"Benchmarking Geospatial Question Answering Engines using the Dataset GeoQuestions1089","date":"2023-11-06","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/chatkbqa-a-generate-then-retrieve-framework","title":"ChatKBQA: A Generate-then-Retrieve Framework for Knowledge Base Question Answering with Fine-tuned Large Language Models","date":"2023-10-13","arxiv_id":"2310.08975","repositories_listed":1,"syntology":{"n":16,"n_ran":8,"n_unverified":8,"n_pointer_only":0}},{"url":"/paper/data-distribution-bottlenecks-in-grounding","title":"Data Distribution Bottlenecks in Grounding Language Models to Knowledge Bases","date":"2023-09-15","arxiv_id":"2309.08345","repositories_listed":1,"syntology":null},{"url":"/paper/bridging-the-kb-text-gap-leveraging","title":"Bridging the KB-Text Gap: Leveraging Structured Knowledge-aware Pre-training for KBQA","date":"2023-08-28","arxiv_id":"2308.14436","repositories_listed":1,"syntology":{"n":10,"n_ran":10,"n_unverified":0,"n_pointer_only":10}},{"url":"/paper/flexkbqa-a-flexible-llm-powered-framework-for","title":"FlexKBQA: A Flexible LLM-Powered Framework for Few-Shot Knowledge Base Question Answering","date":"2023-08-23","arxiv_id":"2308.12060","repositories_listed":1,"syntology":{"n":4,"n_ran":4,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/question-answering-over-linked-data-with-gpt","title":"Question Answering over Linked Data with GPT-3","date":"2023-08-15","arxiv_id":null,"repositories_listed":1,"syntology":null},{"url":"/paper/question-decomposition-tree-for-answering","title":"Question Decomposition Tree for Answering Complex Questions over Knowledge Bases","date":"2023-06-13","arxiv_id":"2306.07597","repositories_listed":1,"syntology":null},{"url":"/paper/knowledge-base-question-answering-for-space","title":"Knowledge Base Question Answering for Space Debris Queries","date":"2023-05-31","arxiv_id":"2305.19734","repositories_listed":1,"syntology":null},{"url":"/paper/complementing-gpt-3-with-few-shot-sequence-to","title":"Fine-tuned LLMs Know More, Hallucinate Less with Few-Shot Sequence-to-Sequence Semantic Parsing over Wikidata","date":"2023-05-23","arxiv_id":"2305.14202","repositories_listed":1,"syntology":null}],"syntology_records":11,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}