{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2026:AOKQ6YYVUQZPFFQUNQF4EGYLGY","short_pith_number":"pith:AOKQ6YYV","schema_version":"1.0","canonical_sha256":"03950f6315a432f296146c0bc21b0b36378d60e8ca37876ba199466c45b6347d","source":{"kind":"arxiv","id":"2608.11694","version":1},"attestation_state":"computed","paper":{"title":"The Wording Effect: Quantifying Two-Way Drift in LLM Benchmark Performance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chad DeLuca, Hima Patel, Shailja Thakur, Sungeun An","submitted_at":"2026-08-12T06:05:47Z","abstract_excerpt":"A benchmark score comes from a single phrasing of each problem. That single phrasing is treated as if it stood for the whole space of ways the same problem could be asked, but it does not. We show that rephrasing a problem while keeping its meaning and answer fixed routinely flips a model's answer in both directions, so some failures become successes and some successes become failures. We call this drift. BenchDrift generates meaning-preserving variations of benchmark problems along four axes, namely linguistic, referential, pragmatic, and structural, and measures how often, and why, correctne"},"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":"2608.11694","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-08-12T06:05:47Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"21daa4f73bfaaa54d4e27742301e6f3e2bfcffd14258f06583414bd396e9ec4f","abstract_canon_sha256":"9faf026f12361cfc7e298e1c3c06cd851ba40b9724107217b30df313f697a23a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-08-13T01:27:51.401837Z","signature_b64":"mZRut07H+XQGB2SgAS4E1NJSwPxu0YxK0PrRCPuVHMW/LQyMSwr7fuXwy8E4ppcuoFjkMZFF8N0Vy6LyrmjjCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"03950f6315a432f296146c0bc21b0b36378d60e8ca37876ba199466c45b6347d","last_reissued_at":"2026-08-13T01:27:51.398419Z","signature_status":"signed_v1","first_computed_at":"2026-08-13T01:27:51.398419Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Wording Effect: Quantifying Two-Way Drift in LLM Benchmark Performance","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Chad DeLuca, Hima Patel, Shailja Thakur, Sungeun An","submitted_at":"2026-08-12T06:05:47Z","abstract_excerpt":"A benchmark score comes from a single phrasing of each problem. That single phrasing is treated as if it stood for the whole space of ways the same problem could be asked, but it does not. We show that rephrasing a problem while keeping its meaning and answer fixed routinely flips a model's answer in both directions, so some failures become successes and some successes become failures. We call this drift. BenchDrift generates meaning-preserving variations of benchmark problems along four axes, namely linguistic, referential, pragmatic, and structural, and measures how often, and why, correctne"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2608.11694","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/2608.11694/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":"2608.11694","created_at":"2026-08-13T01:27:51.400783+00:00"},{"alias_kind":"arxiv_version","alias_value":"2608.11694v1","created_at":"2026-08-13T01:27:51.400783+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2608.11694","created_at":"2026-08-13T01:27:51.400783+00:00"},{"alias_kind":"pith_short_12","alias_value":"AOKQ6YYVUQZP","created_at":"2026-08-13T01:27:51.400783+00:00"},{"alias_kind":"pith_short_16","alias_value":"AOKQ6YYVUQZPFFQU","created_at":"2026-08-13T01:27:51.400783+00:00"},{"alias_kind":"pith_short_8","alias_value":"AOKQ6YYV","created_at":"2026-08-13T01:27:51.400783+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/AOKQ6YYVUQZPFFQUNQF4EGYLGY","json":"https://pith.science/pith/AOKQ6YYVUQZPFFQUNQF4EGYLGY.json","graph_json":"https://pith.science/api/pith-number/AOKQ6YYVUQZPFFQUNQF4EGYLGY/graph.json","events_json":"https://pith.science/api/pith-number/AOKQ6YYVUQZPFFQUNQF4EGYLGY/events.json","paper":"https://pith.science/paper/AOKQ6YYV"},"agent_actions":{"view_html":"https://pith.science/pith/AOKQ6YYVUQZPFFQUNQF4EGYLGY","download_json":"https://pith.science/pith/AOKQ6YYVUQZPFFQUNQF4EGYLGY.json","view_paper":"https://pith.science/paper/AOKQ6YYV","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2608.11694&json=true","fetch_graph":"https://pith.science/api/pith-number/AOKQ6YYVUQZPFFQUNQF4EGYLGY/graph.json","fetch_events":"https://pith.science/api/pith-number/AOKQ6YYVUQZPFFQUNQF4EGYLGY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/AOKQ6YYVUQZPFFQUNQF4EGYLGY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/AOKQ6YYVUQZPFFQUNQF4EGYLGY/action/storage_attestation","attest_author":"https://pith.science/pith/AOKQ6YYVUQZPFFQUNQF4EGYLGY/action/author_attestation","sign_citation":"https://pith.science/pith/AOKQ6YYVUQZPFFQUNQF4EGYLGY/action/citation_signature","submit_replication":"https://pith.science/pith/AOKQ6YYVUQZPFFQUNQF4EGYLGY/action/replication_record"}},"created_at":"2026-08-13T01:27:51.400783+00:00","updated_at":"2026-08-13T01:27:51.400783+00:00"}