{"id":13605472,"url":"https://github.com/Akirato/LLM-KG-Reasoning","last_synced_at":"2025-04-12T05:33:14.939Z","repository":{"id":160182672,"uuid":"621572317","full_name":"Akirato/LLM-KG-Reasoning","owner":"Akirato","description":"We want to try and evaluate LLMs using Knowledge Graphs","archived":false,"fork":false,"pushed_at":"2023-05-02T01:57:59.000Z","size":623,"stargazers_count":101,"open_issues_count":5,"forks_count":9,"subscribers_count":6,"default_branch":"main","last_synced_at":"2024-10-09T10:05:54.874Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"gpl-3.0","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/Akirato.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null}},"created_at":"2023-03-31T00:02:12.000Z","updated_at":"2024-10-08T03:43:54.000Z","dependencies_parsed_at":"2023-05-23T13:46:29.264Z","dependency_job_id":null,"html_url":"https://github.com/Akirato/LLM-KG-Reasoning","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Akirato%2FLLM-KG-Reasoning","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Akirato%2FLLM-KG-Reasoning/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Akirato%2FLLM-KG-Reasoning/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Akirato%2FLLM-KG-Reasoning/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Akirato","download_url":"https://codeload.github.com/Akirato/LLM-KG-Reasoning/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":223497867,"owners_count":17155214,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":[],"created_at":"2024-08-01T19:00:59.091Z","updated_at":"2024-11-07T10:30:39.912Z","avatar_url":"https://github.com/Akirato.png","language":"Python","funding_links":[],"categories":[":wrench: Implementations"],"sub_categories":["Dataset tools"],"readme":"# Reasoning over Knowledge Graphs using Large Language Models\n![Overview of LARK model.](./assets/model.png)\n### Abstract\nReasoning over knowledge graphs (KGs) is a challenging task that requires a deep\nunderstanding of the complex relationships between entities and the underlying\nlogic of their relations. Current approaches rely on learning geometries to embed\nentities in vector space for logical query operations, but they suffer from subpar\nperformance on complex queries and dataset-specific representations. In this paper,\nwe propose a novel decoupled approach, Language-guided Abstract Reasoning\nover Knowledge graphs (LARK), that formulates complex KG reasoning as a\ncombination of contextual KG search and abstract logical query reasoning, to\nleverage the strengths of graph extraction algorithms and large language models\n(LLM), respectively. Our experiments demonstrate that the proposed approach\noutperforms state-of-the-art KG reasoning methods on standard benchmark datasets\nacross several logical query constructs, with significant performance gain for\nqueries of higher complexity. Furthermore, we show that the performance of our\napproach improves proportionally to the increase in size of the underlying LLM,\nenabling the integration of the latest advancements in LLMs for logical reasoning\nover KGs. Our work presents a new direction for addressing the challenges of\ncomplex KG reasoning and paves the way for future research in this area.\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FAkirato%2FLLM-KG-Reasoning","html_url":"https://awesome.ecosyste.ms/projects/github.com%2FAkirato%2FLLM-KG-Reasoning","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2FAkirato%2FLLM-KG-Reasoning/lists"}