{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2024:4ARFAADXWP5FKXWOKPAMLPYGL4","short_pith_number":"pith:4ARFAADX","schema_version":"1.0","canonical_sha256":"e022500077b3fa555ece53c0c5bf065f2eaef2841f548b99038f7ab8c28305a2","source":{"kind":"arxiv","id":"2407.12830","version":3},"attestation_state":"computed","paper":{"title":"Knowledge-based Consistency Testing of Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ezekiel Soremekun, Sai Sathiesh Rajan, Sudipta Chattopadhyay","submitted_at":"2024-07-03T11:16:54Z","abstract_excerpt":"In this work, we systematically expose and measure the inconsistency and knowledge gaps of Large Language Models (LLMs). Specifically, we propose an automated testing framework (called KonTest) which leverages a knowledge graph to construct test cases. KonTest probes and measures the inconsistencies in the LLM's knowledge of the world via a combination of semantically-equivalent queries and test oracles (metamorphic or ontological oracle). KonTest further mitigates knowledge gaps via a weighted LLM model ensemble. Using four state-of-the-art LLMs (Falcon, Gemini, GPT3.5, and Llama2), we show t"},"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":"2407.12830","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-03T11:16:54Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"c884685f03991625618de2ce812cbaa42a0415106c7e8755391221f460bb20c9","abstract_canon_sha256":"10b1f089478337116d82c79c8446fee09b5afe2c837277bff8ed566b11c7bf0b"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:53:41.191880Z","signature_b64":"NoBywhLMGgpSgbQsIgRo5rZX6EHsJmPFjVagoSZx3q5gIXWr139I8wIpOwR/3aUY/nPAcBTqcUk0C87hDqCzAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e022500077b3fa555ece53c0c5bf065f2eaef2841f548b99038f7ab8c28305a2","last_reissued_at":"2026-07-05T11:53:41.191393Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:53:41.191393Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Knowledge-based Consistency Testing of Large Language Models","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CL","authors_text":"Ezekiel Soremekun, Sai Sathiesh Rajan, Sudipta Chattopadhyay","submitted_at":"2024-07-03T11:16:54Z","abstract_excerpt":"In this work, we systematically expose and measure the inconsistency and knowledge gaps of Large Language Models (LLMs). Specifically, we propose an automated testing framework (called KonTest) which leverages a knowledge graph to construct test cases. KonTest probes and measures the inconsistencies in the LLM's knowledge of the world via a combination of semantically-equivalent queries and test oracles (metamorphic or ontological oracle). KonTest further mitigates knowledge gaps via a weighted LLM model ensemble. Using four state-of-the-art LLMs (Falcon, Gemini, GPT3.5, and Llama2), we show t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.12830","kind":"arxiv","version":3},"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/2407.12830/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":"2407.12830","created_at":"2026-07-05T11:53:41.191450+00:00"},{"alias_kind":"arxiv_version","alias_value":"2407.12830v3","created_at":"2026-07-05T11:53:41.191450+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.12830","created_at":"2026-07-05T11:53:41.191450+00:00"},{"alias_kind":"pith_short_12","alias_value":"4ARFAADXWP5F","created_at":"2026-07-05T11:53:41.191450+00:00"},{"alias_kind":"pith_short_16","alias_value":"4ARFAADXWP5FKXWO","created_at":"2026-07-05T11:53:41.191450+00:00"},{"alias_kind":"pith_short_8","alias_value":"4ARFAADX","created_at":"2026-07-05T11:53:41.191450+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2506.02546","citing_title":"To trust or not to trust: Attention-based Trust Management for LLM Multi-Agent Systems","ref_index":18,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4ARFAADXWP5FKXWOKPAMLPYGL4","json":"https://pith.science/pith/4ARFAADXWP5FKXWOKPAMLPYGL4.json","graph_json":"https://pith.science/api/pith-number/4ARFAADXWP5FKXWOKPAMLPYGL4/graph.json","events_json":"https://pith.science/api/pith-number/4ARFAADXWP5FKXWOKPAMLPYGL4/events.json","paper":"https://pith.science/paper/4ARFAADX"},"agent_actions":{"view_html":"https://pith.science/pith/4ARFAADXWP5FKXWOKPAMLPYGL4","download_json":"https://pith.science/pith/4ARFAADXWP5FKXWOKPAMLPYGL4.json","view_paper":"https://pith.science/paper/4ARFAADX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2407.12830&json=true","fetch_graph":"https://pith.science/api/pith-number/4ARFAADXWP5FKXWOKPAMLPYGL4/graph.json","fetch_events":"https://pith.science/api/pith-number/4ARFAADXWP5FKXWOKPAMLPYGL4/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4ARFAADXWP5FKXWOKPAMLPYGL4/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4ARFAADXWP5FKXWOKPAMLPYGL4/action/storage_attestation","attest_author":"https://pith.science/pith/4ARFAADXWP5FKXWOKPAMLPYGL4/action/author_attestation","sign_citation":"https://pith.science/pith/4ARFAADXWP5FKXWOKPAMLPYGL4/action/citation_signature","submit_replication":"https://pith.science/pith/4ARFAADXWP5FKXWOKPAMLPYGL4/action/replication_record"}},"created_at":"2026-07-05T11:53:41.191450+00:00","updated_at":"2026-07-05T11:53:41.191450+00:00"}