{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:LTJFUCK5ZM3BSYLOHFN27E7Y3B","short_pith_number":"pith:LTJFUCK5","schema_version":"1.0","canonical_sha256":"5cd25a095dcb3619616e395baf93f8d85686a1934ef7ab16852a8bd953a7b55b","source":{"kind":"arxiv","id":"2605.27268","version":1},"attestation_state":"computed","paper":{"title":"Lost in Sampling: Assessing Lexical Reachability in LLMs via the Word Coverage Score (WCS)","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Carlos Arriaga, Javier Conde, Javier Coronado-Bl\\'azquez, Pedro Reviriego, Samer Awad, Tairan Fu","submitted_at":"2026-05-26T16:44:25Z","abstract_excerpt":"Modern Large Language Models (LLMs) are often criticized for producing repetitive and homogeneous text, despite possessing vast latent vocabularies. While previous research has focused on model knowledge and training data, we investigate the role of decoding mechanics in suppressing linguistic diversity. We introduce the Word Coverage Score (WCS), a metric that quantifies the extent to which contextually appropriate human vocabulary is mathematically pruned by standard sampling filters (e.g., Top-$p$, Top-$k$, and Min-$p$). Rather than assessing static knowledge, the WCS measures the lexical s"},"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":"2605.27268","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-26T16:44:25Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4f565736fe728e0535c9a20da4d4138170dcf4fd4cd8a4fea40366c34c9af145","abstract_canon_sha256":"c2454c63f4ce3958b128f4ffa96e718cffcfc978869d3dbcb7320b6c4052d0f1"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-27T02:06:09.121158Z","signature_b64":"T5F0DKVr4VM0kmoXxySlsNkQ0K6s+PDGq+c5rvChyzMktZvxBJ+6jw6xkeqNIDQbGmlJbJZqkp6jgcqbYLN9Dw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5cd25a095dcb3619616e395baf93f8d85686a1934ef7ab16852a8bd953a7b55b","last_reissued_at":"2026-05-27T02:06:09.120688Z","signature_status":"signed_v1","first_computed_at":"2026-05-27T02:06:09.120688Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Lost in Sampling: Assessing Lexical Reachability in LLMs via the Word Coverage Score (WCS)","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Carlos Arriaga, Javier Conde, Javier Coronado-Bl\\'azquez, Pedro Reviriego, Samer Awad, Tairan Fu","submitted_at":"2026-05-26T16:44:25Z","abstract_excerpt":"Modern Large Language Models (LLMs) are often criticized for producing repetitive and homogeneous text, despite possessing vast latent vocabularies. While previous research has focused on model knowledge and training data, we investigate the role of decoding mechanics in suppressing linguistic diversity. We introduce the Word Coverage Score (WCS), a metric that quantifies the extent to which contextually appropriate human vocabulary is mathematically pruned by standard sampling filters (e.g., Top-$p$, Top-$k$, and Min-$p$). Rather than assessing static knowledge, the WCS measures the lexical s"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.27268","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/2605.27268/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":"2605.27268","created_at":"2026-05-27T02:06:09.120758+00:00"},{"alias_kind":"arxiv_version","alias_value":"2605.27268v1","created_at":"2026-05-27T02:06:09.120758+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.27268","created_at":"2026-05-27T02:06:09.120758+00:00"},{"alias_kind":"pith_short_12","alias_value":"LTJFUCK5ZM3B","created_at":"2026-05-27T02:06:09.120758+00:00"},{"alias_kind":"pith_short_16","alias_value":"LTJFUCK5ZM3BSYLO","created_at":"2026-05-27T02:06:09.120758+00:00"},{"alias_kind":"pith_short_8","alias_value":"LTJFUCK5","created_at":"2026-05-27T02:06:09.120758+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/LTJFUCK5ZM3BSYLOHFN27E7Y3B","json":"https://pith.science/pith/LTJFUCK5ZM3BSYLOHFN27E7Y3B.json","graph_json":"https://pith.science/api/pith-number/LTJFUCK5ZM3BSYLOHFN27E7Y3B/graph.json","events_json":"https://pith.science/api/pith-number/LTJFUCK5ZM3BSYLOHFN27E7Y3B/events.json","paper":"https://pith.science/paper/LTJFUCK5"},"agent_actions":{"view_html":"https://pith.science/pith/LTJFUCK5ZM3BSYLOHFN27E7Y3B","download_json":"https://pith.science/pith/LTJFUCK5ZM3BSYLOHFN27E7Y3B.json","view_paper":"https://pith.science/paper/LTJFUCK5","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2605.27268&json=true","fetch_graph":"https://pith.science/api/pith-number/LTJFUCK5ZM3BSYLOHFN27E7Y3B/graph.json","fetch_events":"https://pith.science/api/pith-number/LTJFUCK5ZM3BSYLOHFN27E7Y3B/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/LTJFUCK5ZM3BSYLOHFN27E7Y3B/action/timestamp_anchor","attest_storage":"https://pith.science/pith/LTJFUCK5ZM3BSYLOHFN27E7Y3B/action/storage_attestation","attest_author":"https://pith.science/pith/LTJFUCK5ZM3BSYLOHFN27E7Y3B/action/author_attestation","sign_citation":"https://pith.science/pith/LTJFUCK5ZM3BSYLOHFN27E7Y3B/action/citation_signature","submit_replication":"https://pith.science/pith/LTJFUCK5ZM3BSYLOHFN27E7Y3B/action/replication_record"}},"created_at":"2026-05-27T02:06:09.120758+00:00","updated_at":"2026-05-27T02:06:09.120758+00:00"}