{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:ZQOCIJP732PIORM2G4BZDONX6H","short_pith_number":"pith:ZQOCIJP7","schema_version":"1.0","canonical_sha256":"cc1c2425ffde9e87459a370391b9b7f1d00965c16b505daf753601e7a07a1f1f","source":{"kind":"arxiv","id":"2607.14112","version":1},"attestation_state":"computed","paper":{"title":"Information-Theoretic Limits of Reliability and Scaling in Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.IT","math.IT"],"primary_cat":"cs.CL","authors_text":"Subhabrata Majumdar","submitted_at":"2026-05-08T06:06:18Z","abstract_excerpt":"Large language models (LLMs) are evaluated as though perfect reliability is achievable for any task given sufficient scale. We show this assumption is information-theoretically unjustified. Every generative task has a reliability ceiling that no model can exceed, determined by how much output uncertainty is resolvable from observable context. The gap decomposes into a resolvable component closable with additional context and a subjective component inherent to task ambiguity. Autoregressive generation further degrades this ceiling at a rate governed by the task's dependency kernel, which quanti"},"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":"2607.14112","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-08T06:06:18Z","cross_cats_sorted":["cs.AI","cs.IT","math.IT"],"title_canon_sha256":"d0cc6fed0e3cd3961e28933c667f7fc379dde282c19f52a760e3647733eb679e","abstract_canon_sha256":"b5e2be4c1313a6b6fde197c679109f095c5ee4d79ec496b0651cb4c9997ed263"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-17T00:20:52.558581Z","signature_b64":"xkVQ0XGu9w/87pkBVkPM0fqqWCVKRz1976wPSWNbmCTABsAJlkx6PHp+saLl7T5CjrMira4g7c/IKrfECn09DA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"cc1c2425ffde9e87459a370391b9b7f1d00965c16b505daf753601e7a07a1f1f","last_reissued_at":"2026-07-17T00:20:52.557762Z","signature_status":"signed_v1","first_computed_at":"2026-07-17T00:20:52.557762Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Information-Theoretic Limits of Reliability and Scaling in Language Models","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.IT","math.IT"],"primary_cat":"cs.CL","authors_text":"Subhabrata Majumdar","submitted_at":"2026-05-08T06:06:18Z","abstract_excerpt":"Large language models (LLMs) are evaluated as though perfect reliability is achievable for any task given sufficient scale. We show this assumption is information-theoretically unjustified. Every generative task has a reliability ceiling that no model can exceed, determined by how much output uncertainty is resolvable from observable context. The gap decomposes into a resolvable component closable with additional context and a subjective component inherent to task ambiguity. Autoregressive generation further degrades this ceiling at a rate governed by the task's dependency kernel, which quanti"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2607.14112","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/2607.14112/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":"2607.14112","created_at":"2026-07-17T00:20:52.558185+00:00"},{"alias_kind":"arxiv_version","alias_value":"2607.14112v1","created_at":"2026-07-17T00:20:52.558185+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2607.14112","created_at":"2026-07-17T00:20:52.558185+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZQOCIJP732PI","created_at":"2026-07-17T00:20:52.558185+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZQOCIJP732PIORM2","created_at":"2026-07-17T00:20:52.558185+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZQOCIJP7","created_at":"2026-07-17T00:20:52.558185+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/ZQOCIJP732PIORM2G4BZDONX6H","json":"https://pith.science/pith/ZQOCIJP732PIORM2G4BZDONX6H.json","graph_json":"https://pith.science/api/pith-number/ZQOCIJP732PIORM2G4BZDONX6H/graph.json","events_json":"https://pith.science/api/pith-number/ZQOCIJP732PIORM2G4BZDONX6H/events.json","paper":"https://pith.science/paper/ZQOCIJP7"},"agent_actions":{"view_html":"https://pith.science/pith/ZQOCIJP732PIORM2G4BZDONX6H","download_json":"https://pith.science/pith/ZQOCIJP732PIORM2G4BZDONX6H.json","view_paper":"https://pith.science/paper/ZQOCIJP7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2607.14112&json=true","fetch_graph":"https://pith.science/api/pith-number/ZQOCIJP732PIORM2G4BZDONX6H/graph.json","fetch_events":"https://pith.science/api/pith-number/ZQOCIJP732PIORM2G4BZDONX6H/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZQOCIJP732PIORM2G4BZDONX6H/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZQOCIJP732PIORM2G4BZDONX6H/action/storage_attestation","attest_author":"https://pith.science/pith/ZQOCIJP732PIORM2G4BZDONX6H/action/author_attestation","sign_citation":"https://pith.science/pith/ZQOCIJP732PIORM2G4BZDONX6H/action/citation_signature","submit_replication":"https://pith.science/pith/ZQOCIJP732PIORM2G4BZDONX6H/action/replication_record"}},"created_at":"2026-07-17T00:20:52.558185+00:00","updated_at":"2026-07-17T00:20:52.558185+00:00"}