{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LJQICMEQGX7H4LAB5CBHWLI2QK","short_pith_number":"pith:LJQICMEQ","schema_version":"1.0","canonical_sha256":"5a6081309035fe7e2c01e8827b2d1a8299ea9e9f0345f26e0b2aa149578af5af","source":{"kind":"arxiv","id":"2606.13649","version":1},"attestation_state":"computed","paper":{"title":"Operadic consistency: a label-free signal for compositional reasoning failures in LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Kyle Richardson, Nathaniel Bottman, Yinhong Liu","submitted_at":"2026-06-11T17:50:40Z","abstract_excerpt":"Detecting LLM reasoning failures at inference time without ground-truth labels has motivated a wide range of confidence baselines, including self-consistency, semantic entropy, and P(True), built on within-question sampling and self-evaluation. Operad theory, the formalism for systems built by iterated substitution, suggests a complementary diagnostic: a model's direct answer to a compositional query should agree with the answer it produces by composing a stated decomposition of the same query. We instantiate this idea as operadic consistency (OC), a per-question signal. Across twelve instruct"},"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":"2606.13649","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-06-11T17:50:40Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"0a192e16502b0700ea5c1727b5b215530699fc3e2e7fb4f1a9d58838f2b319cc","abstract_canon_sha256":"17bd1cf83e44305255df031a6dafcd909a12e9e9976144d8a9b8594143625424"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-06-12T01:10:21.043427Z","signature_b64":"l5mREPZzGa+Td3qZanPdLNrtnezJ+WavMjHdjg8f4ZFL1m5TNgSgum4/OnxxdSn5SIKQKqCmUBRkx0RjKrYDBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5a6081309035fe7e2c01e8827b2d1a8299ea9e9f0345f26e0b2aa149578af5af","last_reissued_at":"2026-06-12T01:10:21.042595Z","signature_status":"signed_v1","first_computed_at":"2026-06-12T01:10:21.042595Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Operadic consistency: a label-free signal for compositional reasoning failures in LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"cs.CL","authors_text":"Kyle Richardson, Nathaniel Bottman, Yinhong Liu","submitted_at":"2026-06-11T17:50:40Z","abstract_excerpt":"Detecting LLM reasoning failures at inference time without ground-truth labels has motivated a wide range of confidence baselines, including self-consistency, semantic entropy, and P(True), built on within-question sampling and self-evaluation. Operad theory, the formalism for systems built by iterated substitution, suggests a complementary diagnostic: a model's direct answer to a compositional query should agree with the answer it produces by composing a stated decomposition of the same query. We instantiate this idea as operadic consistency (OC), a per-question signal. Across twelve instruct"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2606.13649","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/2606.13649/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":"2606.13649","created_at":"2026-06-12T01:10:21.042718+00:00"},{"alias_kind":"arxiv_version","alias_value":"2606.13649v1","created_at":"2026-06-12T01:10:21.042718+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2606.13649","created_at":"2026-06-12T01:10:21.042718+00:00"},{"alias_kind":"pith_short_12","alias_value":"LJQICMEQGX7H","created_at":"2026-06-12T01:10:21.042718+00:00"},{"alias_kind":"pith_short_16","alias_value":"LJQICMEQGX7H4LAB","created_at":"2026-06-12T01:10:21.042718+00:00"},{"alias_kind":"pith_short_8","alias_value":"LJQICMEQ","created_at":"2026-06-12T01:10:21.042718+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/LJQICMEQGX7H4LAB5CBHWLI2QK","json":"https://pith.science/pith/LJQICMEQGX7H4LAB5CBHWLI2QK.json","graph_json":"https://pith.science/api/pith-number/LJQICMEQGX7H4LAB5CBHWLI2QK/graph.json","events_json":"https://pith.science/api/pith-number/LJQICMEQGX7H4LAB5CBHWLI2QK/events.json","paper":"https://pith.science/paper/LJQICMEQ"},"agent_actions":{"view_html":"https://pith.science/pith/LJQICMEQGX7H4LAB5CBHWLI2QK","download_json":"https://pith.science/pith/LJQICMEQGX7H4LAB5CBHWLI2QK.json","view_paper":"https://pith.science/paper/LJQICMEQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2606.13649&json=true","fetch_graph":"https://pith.science/api/pith-number/LJQICMEQGX7H4LAB5CBHWLI2QK/graph.json","fetch_events":"https://pith.science/api/pith-number/LJQICMEQGX7H4LAB5CBHWLI2QK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LJQICMEQGX7H4LAB5CBHWLI2QK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LJQICMEQGX7H4LAB5CBHWLI2QK/action/storage_attestation","attest_author":"https://pith.science/pith/LJQICMEQGX7H4LAB5CBHWLI2QK/action/author_attestation","sign_citation":"https://pith.science/pith/LJQICMEQGX7H4LAB5CBHWLI2QK/action/citation_signature","submit_replication":"https://pith.science/pith/LJQICMEQGX7H4LAB5CBHWLI2QK/action/replication_record"}},"created_at":"2026-06-12T01:10:21.042718+00:00","updated_at":"2026-06-12T01:10:21.042718+00:00"}