{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:X6PVUDIBM5FPUKCVUX4Z2UDAZA","short_pith_number":"pith:X6PVUDIB","schema_version":"1.0","canonical_sha256":"bf9f5a0d01674afa2855a5f99d5060c80483ad6b966c5c53e0fe754ab2fe4efe","source":{"kind":"arxiv","id":"2605.24850","version":1},"attestation_state":"computed","paper":{"title":"Repeated Sequences Reveal Gaps between Large Language Models and Natural Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IT","math.IT","stat.AP"],"primary_cat":"cs.CL","authors_text":"Kumiko Tanaka-Ishii","submitted_at":"2026-05-24T03:49:20Z","abstract_excerpt":"Evaluating whether large language models (LLMs) capture the structure of natural language beyond local fluency remains an open challenge. Existing evaluation methods, largely based on task performance or short-context behavior, provide limited insight into the long-range statistical organization of generated text.\n  We propose a complementary evaluation framework based on repeated subsequences. By analyzing their distribution across scales and relating it to higher-order R\\'enyi entropies, we probe how texts reuse previously established structure under finite-length conditions. Experiments on "},"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":"2605.24850","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-24T03:49:20Z","cross_cats_sorted":["cs.IT","math.IT","stat.AP"],"title_canon_sha256":"211d100cd92d7651a25ffe8c7ac82915e773874a2a39c30397dde974b6d32395","abstract_canon_sha256":"4ed3fb40d3b38964031be9b5a90e804c183918154bde5ed5ac8eac6e1943d7f8"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-26T01:04:01.670222Z","signature_b64":"jDOylPtUPvuCSUwPGlL2V5bL1zbgjVejYGyUJQoD2EJE0XuL/9qxqSfS0SFmEbh/L+U173//VvOWaMa4MNGgAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bf9f5a0d01674afa2855a5f99d5060c80483ad6b966c5c53e0fe754ab2fe4efe","last_reissued_at":"2026-05-26T01:04:01.669480Z","signature_status":"signed_v1","first_computed_at":"2026-05-26T01:04:01.669480Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Repeated Sequences Reveal Gaps between Large Language Models and Natural Language","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IT","math.IT","stat.AP"],"primary_cat":"cs.CL","authors_text":"Kumiko Tanaka-Ishii","submitted_at":"2026-05-24T03:49:20Z","abstract_excerpt":"Evaluating whether large language models (LLMs) capture the structure of natural language beyond local fluency remains an open challenge. Existing evaluation methods, largely based on task performance or short-context behavior, provide limited insight into the long-range statistical organization of generated text.\n  We propose a complementary evaluation framework based on repeated subsequences. By analyzing their distribution across scales and relating it to higher-order R\\'enyi entropies, we probe how texts reuse previously established structure under finite-length conditions. Experiments on "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.24850","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/2605.24850/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":"2605.24850","created_at":"2026-05-26T01:04:01.669594+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.24850v1","created_at":"2026-05-26T01:04:01.669594+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.24850","created_at":"2026-05-26T01:04:01.669594+00:00"},{"alias_kind":"pith_short_12","alias_value":"X6PVUDIBM5FP","created_at":"2026-05-26T01:04:01.669594+00:00"},{"alias_kind":"pith_short_16","alias_value":"X6PVUDIBM5FPUKCV","created_at":"2026-05-26T01:04:01.669594+00:00"},{"alias_kind":"pith_short_8","alias_value":"X6PVUDIB","created_at":"2026-05-26T01:04:01.669594+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2605.00435","citing_title":"Escaping Mode Collapse in LLM Generation via Geometric Regulation","ref_index":40,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/X6PVUDIBM5FPUKCVUX4Z2UDAZA","json":"https://pith.science/pith/X6PVUDIBM5FPUKCVUX4Z2UDAZA.json","graph_json":"https://pith.science/api/pith-number/X6PVUDIBM5FPUKCVUX4Z2UDAZA/graph.json","events_json":"https://pith.science/api/pith-number/X6PVUDIBM5FPUKCVUX4Z2UDAZA/events.json","paper":"https://pith.science/paper/X6PVUDIB"},"agent_actions":{"view_html":"https://pith.science/pith/X6PVUDIBM5FPUKCVUX4Z2UDAZA","download_json":"https://pith.science/pith/X6PVUDIBM5FPUKCVUX4Z2UDAZA.json","view_paper":"https://pith.science/paper/X6PVUDIB","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.24850&json=true","fetch_graph":"https://pith.science/api/pith-number/X6PVUDIBM5FPUKCVUX4Z2UDAZA/graph.json","fetch_events":"https://pith.science/api/pith-number/X6PVUDIBM5FPUKCVUX4Z2UDAZA/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/X6PVUDIBM5FPUKCVUX4Z2UDAZA/action/timestamp_anchor","attest_storage":"https://pith.science/pith/X6PVUDIBM5FPUKCVUX4Z2UDAZA/action/storage_attestation","attest_author":"https://pith.science/pith/X6PVUDIBM5FPUKCVUX4Z2UDAZA/action/author_attestation","sign_citation":"https://pith.science/pith/X6PVUDIBM5FPUKCVUX4Z2UDAZA/action/citation_signature","submit_replication":"https://pith.science/pith/X6PVUDIBM5FPUKCVUX4Z2UDAZA/action/replication_record"}},"created_at":"2026-05-26T01:04:01.669594+00:00","updated_at":"2026-05-26T01:04:01.669594+00:00"}