{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:XKORP3UC2AUQGD62EOYGOXJN6S","short_pith_number":"pith:XKORP3UC","schema_version":"1.0","canonical_sha256":"ba9d17ee82d029030fda23b0675d2df4a424730433aa72604f47ce45d0a45ebf","source":{"kind":"arxiv","id":"2506.10095","version":1},"attestation_state":"computed","paper":{"title":"When Meaning Stays the Same, but Models Drift: Evaluating Quality of Service under Token-Level Behavioral Instability in LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alan Peters, Joel Kreuzwieser, Xiao Li","submitted_at":"2025-06-11T18:26:32Z","abstract_excerpt":"We investigate how large language models respond to prompts that differ only in their token-level realization but preserve the same semantic intent, a phenomenon we call prompt variance. We propose Prompt-Based Semantic Shift (PBSS), a diagnostic framework for measuring behavioral drift in LLMs under semantically equivalent prompt rewordings. Applied to ten constrained tasks, PBSS reveals consistent, model-specific response shifts, suggesting statistical regularities linked to tokenization and decoding. These results highlight an overlooked dimension of model evaluation stability under rephras"},"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":"2506.10095","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T18:26:32Z","cross_cats_sorted":[],"title_canon_sha256":"d811981b35f4e8c2b649c1b5fa5164448ee9764ffdada4d68d58e8a73391b93d","abstract_canon_sha256":"32edcd8df380db51d05639f01ae364b2314bb3d170c0164efba02b807a02d9aa"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:20:16.404198Z","signature_b64":"JLBu8hX3dNBjWhHKMFg7KKjJ3gqmhWATgUAVWtyQhwRQ0fyqh+X73GGi5XG6qwkHZfra8ru096O4DUHgew9WAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ba9d17ee82d029030fda23b0675d2df4a424730433aa72604f47ce45d0a45ebf","last_reissued_at":"2026-07-05T11:20:16.403623Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:20:16.403623Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"When Meaning Stays the Same, but Models Drift: Evaluating Quality of Service under Token-Level Behavioral Instability in LLMs","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Alan Peters, Joel Kreuzwieser, Xiao Li","submitted_at":"2025-06-11T18:26:32Z","abstract_excerpt":"We investigate how large language models respond to prompts that differ only in their token-level realization but preserve the same semantic intent, a phenomenon we call prompt variance. We propose Prompt-Based Semantic Shift (PBSS), a diagnostic framework for measuring behavioral drift in LLMs under semantically equivalent prompt rewordings. Applied to ten constrained tasks, PBSS reveals consistent, model-specific response shifts, suggesting statistical regularities linked to tokenization and decoding. These results highlight an overlooked dimension of model evaluation stability under rephras"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10095","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/2506.10095/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":"2506.10095","created_at":"2026-07-05T11:20:16.403686+00:00"},{"alias_kind":"arxiv_version","alias_value":"2506.10095v1","created_at":"2026-07-05T11:20:16.403686+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10095","created_at":"2026-07-05T11:20:16.403686+00:00"},{"alias_kind":"pith_short_12","alias_value":"XKORP3UC2AUQ","created_at":"2026-07-05T11:20:16.403686+00:00"},{"alias_kind":"pith_short_16","alias_value":"XKORP3UC2AUQGD62","created_at":"2026-07-05T11:20:16.403686+00:00"},{"alias_kind":"pith_short_8","alias_value":"XKORP3UC","created_at":"2026-07-05T11:20:16.403686+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/XKORP3UC2AUQGD62EOYGOXJN6S","json":"https://pith.science/pith/XKORP3UC2AUQGD62EOYGOXJN6S.json","graph_json":"https://pith.science/api/pith-number/XKORP3UC2AUQGD62EOYGOXJN6S/graph.json","events_json":"https://pith.science/api/pith-number/XKORP3UC2AUQGD62EOYGOXJN6S/events.json","paper":"https://pith.science/paper/XKORP3UC"},"agent_actions":{"view_html":"https://pith.science/pith/XKORP3UC2AUQGD62EOYGOXJN6S","download_json":"https://pith.science/pith/XKORP3UC2AUQGD62EOYGOXJN6S.json","view_paper":"https://pith.science/paper/XKORP3UC","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2506.10095&json=true","fetch_graph":"https://pith.science/api/pith-number/XKORP3UC2AUQGD62EOYGOXJN6S/graph.json","fetch_events":"https://pith.science/api/pith-number/XKORP3UC2AUQGD62EOYGOXJN6S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/XKORP3UC2AUQGD62EOYGOXJN6S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/XKORP3UC2AUQGD62EOYGOXJN6S/action/storage_attestation","attest_author":"https://pith.science/pith/XKORP3UC2AUQGD62EOYGOXJN6S/action/author_attestation","sign_citation":"https://pith.science/pith/XKORP3UC2AUQGD62EOYGOXJN6S/action/citation_signature","submit_replication":"https://pith.science/pith/XKORP3UC2AUQGD62EOYGOXJN6S/action/replication_record"}},"created_at":"2026-07-05T11:20:16.403686+00:00","updated_at":"2026-07-05T11:20:16.403686+00:00"}