{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:F5HWUUQ5O5G5AYSCTKDLAGMMHD","short_pith_number":"pith:F5HWUUQ5","schema_version":"1.0","canonical_sha256":"2f4f6a521d774dd062429a86b0198c38e2ba4ab1b5c25573dde7792c9709ce5d","source":{"kind":"arxiv","id":"2503.22166","version":1},"attestation_state":"computed","paper":{"title":"Reasoning of Large Language Models over Knowledge Graphs with Super-Relations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Julian Shun, Jundong Li, Junhong Lin, Song Wang, Xiaojie Guo, Yada Zhu","submitted_at":"2025-03-28T06:11:04Z","abstract_excerpt":"While large language models (LLMs) have made significant progress in processing and reasoning over knowledge graphs, current methods suffer from a high non-retrieval rate. This limitation reduces the accuracy of answering questions based on these graphs. Our analysis reveals that the combination of greedy search and forward reasoning is a major contributor to this issue. To overcome these challenges, we introduce the concept of super-relations, which enables both forward and backward reasoning by summarizing and connecting various relational paths within the graph. This holistic approach not o"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2503.22166","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-03-28T06:11:04Z","cross_cats_sorted":[],"title_canon_sha256":"73349a5c174585d307bcbabd460b76e84d2963f6e3f2aae8a0610a458bb6006e","abstract_canon_sha256":"8b6cb5dc235c3c7e89609da8bb19f58c6ab37279950e13ca43610c18c0c1ff0b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:40:48.631901Z","signature_b64":"hh3if3J+kjneUxF//fbV7DR5VdW/Xf9Ev7XzaN5ILXWRVea4mrir8TgDRgATGQHcu/GiLbugKu3zKRaNdrDIDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2f4f6a521d774dd062429a86b0198c38e2ba4ab1b5c25573dde7792c9709ce5d","last_reissued_at":"2026-07-05T10:40:48.631428Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:40:48.631428Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Reasoning of Large Language Models over Knowledge Graphs with Super-Relations","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Julian Shun, Jundong Li, Junhong Lin, Song Wang, Xiaojie Guo, Yada Zhu","submitted_at":"2025-03-28T06:11:04Z","abstract_excerpt":"While large language models (LLMs) have made significant progress in processing and reasoning over knowledge graphs, current methods suffer from a high non-retrieval rate. This limitation reduces the accuracy of answering questions based on these graphs. Our analysis reveals that the combination of greedy search and forward reasoning is a major contributor to this issue. To overcome these challenges, we introduce the concept of super-relations, which enables both forward and backward reasoning by summarizing and connecting various relational paths within the graph. This holistic approach not o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2503.22166","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2503.22166/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2503.22166","created_at":"2026-07-05T10:40:48.631485+00:00"},{"alias_kind":"arxiv_version","alias_value":"2503.22166v1","created_at":"2026-07-05T10:40:48.631485+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2503.22166","created_at":"2026-07-05T10:40:48.631485+00:00"},{"alias_kind":"pith_short_12","alias_value":"F5HWUUQ5O5G5","created_at":"2026-07-05T10:40:48.631485+00:00"},{"alias_kind":"pith_short_16","alias_value":"F5HWUUQ5O5G5AYSC","created_at":"2026-07-05T10:40:48.631485+00:00"},{"alias_kind":"pith_short_8","alias_value":"F5HWUUQ5","created_at":"2026-07-05T10:40:48.631485+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2509.26383","citing_title":"Efficient and Transferable Agentic Knowledge Graph RAG via Reinforcement Learning","ref_index":40,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/F5HWUUQ5O5G5AYSCTKDLAGMMHD","json":"https://pith.science/pith/F5HWUUQ5O5G5AYSCTKDLAGMMHD.json","graph_json":"https://pith.science/api/pith-number/F5HWUUQ5O5G5AYSCTKDLAGMMHD/graph.json","events_json":"https://pith.science/api/pith-number/F5HWUUQ5O5G5AYSCTKDLAGMMHD/events.json","paper":"https://pith.science/paper/F5HWUUQ5"},"agent_actions":{"view_html":"https://pith.science/pith/F5HWUUQ5O5G5AYSCTKDLAGMMHD","download_json":"https://pith.science/pith/F5HWUUQ5O5G5AYSCTKDLAGMMHD.json","view_paper":"https://pith.science/paper/F5HWUUQ5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2503.22166&json=true","fetch_graph":"https://pith.science/api/pith-number/F5HWUUQ5O5G5AYSCTKDLAGMMHD/graph.json","fetch_events":"https://pith.science/api/pith-number/F5HWUUQ5O5G5AYSCTKDLAGMMHD/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/F5HWUUQ5O5G5AYSCTKDLAGMMHD/action/timestamp_anchor","attest_storage":"https://pith.science/pith/F5HWUUQ5O5G5AYSCTKDLAGMMHD/action/storage_attestation","attest_author":"https://pith.science/pith/F5HWUUQ5O5G5AYSCTKDLAGMMHD/action/author_attestation","sign_citation":"https://pith.science/pith/F5HWUUQ5O5G5AYSCTKDLAGMMHD/action/citation_signature","submit_replication":"https://pith.science/pith/F5HWUUQ5O5G5AYSCTKDLAGMMHD/action/replication_record"}},"created_at":"2026-07-05T10:40:48.631485+00:00","updated_at":"2026-07-05T10:40:48.631485+00:00"}