{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2025:XKORP3UC2AUQGD62EOYGOXJN6S","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"32edcd8df380db51d05639f01ae364b2314bb3d170c0164efba02b807a02d9aa","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T18:26:32Z","title_canon_sha256":"d811981b35f4e8c2b649c1b5fa5164448ee9764ffdada4d68d58e8a73391b93d"},"schema_version":"1.0","source":{"id":"2506.10095","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2506.10095","created_at":"2026-07-05T11:20:16Z"},{"alias_kind":"arxiv_version","alias_value":"2506.10095v1","created_at":"2026-07-05T11:20:16Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2506.10095","created_at":"2026-07-05T11:20:16Z"},{"alias_kind":"pith_short_12","alias_value":"XKORP3UC2AUQ","created_at":"2026-07-05T11:20:16Z"},{"alias_kind":"pith_short_16","alias_value":"XKORP3UC2AUQGD62","created_at":"2026-07-05T11:20:16Z"},{"alias_kind":"pith_short_8","alias_value":"XKORP3UC","created_at":"2026-07-05T11:20:16Z"}],"graph_snapshots":[{"event_id":"sha256:8113a3bd31d00ff4c71a20cad3eba3fdd41c3d96ff4d537967e0afec45134917","target":"graph","created_at":"2026-07-05T11:20:16Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2506.10095/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"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","authors_text":"Alan Peters, Joel Kreuzwieser, Xiao Li","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T18:26:32Z","title":"When Meaning Stays the Same, but Models Drift: Evaluating Quality of Service under Token-Level Behavioral Instability in LLMs"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2506.10095","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:7f4a96ed59c1206333307aa5c8c5d2bf493a35cbc001b3ab26a3ac4774fe4cb1","target":"record","created_at":"2026-07-05T11:20:16Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"32edcd8df380db51d05639f01ae364b2314bb3d170c0164efba02b807a02d9aa","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2025-06-11T18:26:32Z","title_canon_sha256":"d811981b35f4e8c2b649c1b5fa5164448ee9764ffdada4d68d58e8a73391b93d"},"schema_version":"1.0","source":{"id":"2506.10095","kind":"arxiv","version":1}},"canonical_sha256":"ba9d17ee82d029030fda23b0675d2df4a424730433aa72604f47ce45d0a45ebf","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"ba9d17ee82d029030fda23b0675d2df4a424730433aa72604f47ce45d0a45ebf","first_computed_at":"2026-07-05T11:20:16.403623Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:20:16.403623Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"JLBu8hX3dNBjWhHKMFg7KKjJ3gqmhWATgUAVWtyQhwRQ0fyqh+X73GGi5XG6qwkHZfra8ru096O4DUHgew9WAA==","signature_status":"signed_v1","signed_at":"2026-07-05T11:20:16.404198Z","signed_message":"canonical_sha256_bytes"},"source_id":"2506.10095","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:7f4a96ed59c1206333307aa5c8c5d2bf493a35cbc001b3ab26a3ac4774fe4cb1","sha256:8113a3bd31d00ff4c71a20cad3eba3fdd41c3d96ff4d537967e0afec45134917"],"state_sha256":"82bd0b08b93fbbf0253b0986647155067689475b2f7bd307f1b61bb1127f4d7d"}