{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:FWOG4EGBILFTHWTRHAG5IBLT52","short_pith_number":"pith:FWOG4EGB","schema_version":"1.0","canonical_sha256":"2d9c6e10c142cb33da71380dd40573eebed962da11bb7599c8d6115b9c0f0d1d","source":{"kind":"arxiv","id":"2209.15093","version":1},"attestation_state":"computed","paper":{"title":"Unpacking Large Language Models with Conceptual Consistency","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ajay Divakaran, Michael Cogswell, Pritish Sahu, Yunye Gong","submitted_at":"2022-09-29T20:55:57Z","abstract_excerpt":"If a Large Language Model (LLM) answers \"yes\" to the question \"Are mountains tall?\" then does it know what a mountain is? Can you rely on it responding correctly or incorrectly to other questions about mountains? The success of Large Language Models (LLMs) indicates they are increasingly able to answer queries like these accurately, but that ability does not necessarily imply a general understanding of concepts relevant to the anchor query. We propose conceptual consistency to measure a LLM's understanding of relevant concepts. This novel metric measures how well a model can be characterized b"},"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":"2209.15093","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2022-09-29T20:55:57Z","cross_cats_sorted":[],"title_canon_sha256":"63f7458f9d0ccf2f392b61e4776e99a3def01bef5d2b9c50efaf6950cde55d00","abstract_canon_sha256":"58864a6ad366159538c2ad3f7dd66ffb906937e747367fc1a28a0b344d376956"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:02:15.627100Z","signature_b64":"UndRn0Rv53P7/2/+4fscMhfLLxun+hbgawyQqk+Lgi7XT5VudufxvzJjwlkEb6JVymOjmM3bD+k0Kj9UuPh7CA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"2d9c6e10c142cb33da71380dd40573eebed962da11bb7599c8d6115b9c0f0d1d","last_reissued_at":"2026-07-05T05:02:15.626632Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:02:15.626632Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unpacking Large Language Models with Conceptual Consistency","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Ajay Divakaran, Michael Cogswell, Pritish Sahu, Yunye Gong","submitted_at":"2022-09-29T20:55:57Z","abstract_excerpt":"If a Large Language Model (LLM) answers \"yes\" to the question \"Are mountains tall?\" then does it know what a mountain is? Can you rely on it responding correctly or incorrectly to other questions about mountains? The success of Large Language Models (LLMs) indicates they are increasingly able to answer queries like these accurately, but that ability does not necessarily imply a general understanding of concepts relevant to the anchor query. We propose conceptual consistency to measure a LLM's understanding of relevant concepts. This novel metric measures how well a model can be characterized b"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.15093","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/2209.15093/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":"2209.15093","created_at":"2026-07-05T05:02:15.626689+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.15093v1","created_at":"2026-07-05T05:02:15.626689+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.15093","created_at":"2026-07-05T05:02:15.626689+00:00"},{"alias_kind":"pith_short_12","alias_value":"FWOG4EGBILFT","created_at":"2026-07-05T05:02:15.626689+00:00"},{"alias_kind":"pith_short_16","alias_value":"FWOG4EGBILFTHWTR","created_at":"2026-07-05T05:02:15.626689+00:00"},{"alias_kind":"pith_short_8","alias_value":"FWOG4EGB","created_at":"2026-07-05T05:02:15.626689+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2504.19774","citing_title":"If Concept Bottlenecks are the Question, are Foundation Models the Answer?","ref_index":64,"is_internal_anchor":false},{"citing_arxiv_id":"2605.16405","citing_title":"Concepts Worth Having: Refining VLM-Guided Concept Bottleneck Models with Minimal Annotations","ref_index":1,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/FWOG4EGBILFTHWTRHAG5IBLT52","json":"https://pith.science/pith/FWOG4EGBILFTHWTRHAG5IBLT52.json","graph_json":"https://pith.science/api/pith-number/FWOG4EGBILFTHWTRHAG5IBLT52/graph.json","events_json":"https://pith.science/api/pith-number/FWOG4EGBILFTHWTRHAG5IBLT52/events.json","paper":"https://pith.science/paper/FWOG4EGB"},"agent_actions":{"view_html":"https://pith.science/pith/FWOG4EGBILFTHWTRHAG5IBLT52","download_json":"https://pith.science/pith/FWOG4EGBILFTHWTRHAG5IBLT52.json","view_paper":"https://pith.science/paper/FWOG4EGB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.15093&json=true","fetch_graph":"https://pith.science/api/pith-number/FWOG4EGBILFTHWTRHAG5IBLT52/graph.json","fetch_events":"https://pith.science/api/pith-number/FWOG4EGBILFTHWTRHAG5IBLT52/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/FWOG4EGBILFTHWTRHAG5IBLT52/action/timestamp_anchor","attest_storage":"https://pith.science/pith/FWOG4EGBILFTHWTRHAG5IBLT52/action/storage_attestation","attest_author":"https://pith.science/pith/FWOG4EGBILFTHWTRHAG5IBLT52/action/author_attestation","sign_citation":"https://pith.science/pith/FWOG4EGBILFTHWTRHAG5IBLT52/action/citation_signature","submit_replication":"https://pith.science/pith/FWOG4EGBILFTHWTRHAG5IBLT52/action/replication_record"}},"created_at":"2026-07-05T05:02:15.626689+00:00","updated_at":"2026-07-05T05:02:15.626689+00:00"}