{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2026:Q24F4WTDUUZJBKFTRIHYLH3REB","short_pith_number":"pith:Q24F4WTD","canonical_record":{"source":{"id":"2605.22544","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-21T14:27:46Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"576af77fe98996cd3d6760a49ddcff94fdd18a1544ad9ff8b8792152ee644a68","abstract_canon_sha256":"4e20f71ee5616c5cc507607d492961abe5750987a67f1c05c55ce506af380193"},"schema_version":"1.0"},"canonical_sha256":"86b85e5a63a53290a8b38a0f859f71207df16b4f2642e6762684fdad869f3fba","source":{"kind":"arxiv","id":"2605.22544","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.22544","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.22544v1","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22544","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"pith_short_12","alias_value":"Q24F4WTDUUZJ","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"pith_short_16","alias_value":"Q24F4WTDUUZJBKFT","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"pith_short_8","alias_value":"Q24F4WTD","created_at":"2026-05-22T01:04:56Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2026:Q24F4WTDUUZJBKFTRIHYLH3REB","target":"record","payload":{"canonical_record":{"source":{"id":"2605.22544","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-21T14:27:46Z","cross_cats_sorted":["cs.IR"],"title_canon_sha256":"576af77fe98996cd3d6760a49ddcff94fdd18a1544ad9ff8b8792152ee644a68","abstract_canon_sha256":"4e20f71ee5616c5cc507607d492961abe5750987a67f1c05c55ce506af380193"},"schema_version":"1.0"},"canonical_sha256":"86b85e5a63a53290a8b38a0f859f71207df16b4f2642e6762684fdad869f3fba","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-22T01:04:56.576770Z","signature_b64":"lb3b1tlK+gttWOE2wfiMBRANUFJtXfpBl1iNok68QmS5oCa1geaxRuHdmnUK57BTKfzz9uvPZkWUUe+Kd6W7BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"86b85e5a63a53290a8b38a0f859f71207df16b4f2642e6762684fdad869f3fba","last_reissued_at":"2026-05-22T01:04:56.575939Z","signature_status":"signed_v1","first_computed_at":"2026-05-22T01:04:56.575939Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2605.22544","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:04:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"pAV0o0zZKuWngRkLAeGNH8SFpMwlBLlAzjtzPbNqL7RTOMME7pKZti9deRnlCiV0L6AtMpvhO4LA8U/xkB0oBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T22:39:41.780007Z"},"content_sha256":"e84c5193f11d1e828dc7913f106b8e2cec8355dcd89d9fdc6a7350c19b285c4c","schema_version":"1.0","event_id":"sha256:e84c5193f11d1e828dc7913f106b8e2cec8355dcd89d9fdc6a7350c19b285c4c"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2026:Q24F4WTDUUZJBKFTRIHYLH3REB","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"One prompt is not enough: Instruction Sensitivity Undermines Embedding Model Evaluation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.IR"],"primary_cat":"cs.CL","authors_text":"Kenneth Enevoldsen, Yevhen Kostiuk","submitted_at":"2026-05-21T14:27:46Z","abstract_excerpt":"Instruction embedding models have become common among state-of-the-art models, however are evaluated using a single prompt per task. The single-point evaluation ignores a main problem of the instruction-based approach namely: sensitivity to the phrasing of the instruction. We present an empirical study of prompt sensitivity across 6 embedding models, 11 datasets, and 15 task-specific prompts per dataset, a total of 990. We show that reported scores misrepresent the distribution of scores over plausible prompts. The default prompt can both systematically understate or overstate performance. Fur"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22544","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/2605.22544/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T01:04:56Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"IvIEygPN7cnlz6w2SF1SRzDuim++FgjdWXpPK/U03rPu4HgH7A9ZDHJmJI/QVPHdN+chVRAeDMPzKPKIPFnCBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T22:39:41.780438Z"},"content_sha256":"441f15dfda18647f4bdf42504b51fa4957ae138680d36a88af6a90e91f808d6c","schema_version":"1.0","event_id":"sha256:441f15dfda18647f4bdf42504b51fa4957ae138680d36a88af6a90e91f808d6c"},{"event_type":"integrity_finding","subject_pith_number":"pith:2026:Q24F4WTDUUZJBKFTRIHYLH3REB","target":"integrity","payload":{"note":"DOI in the printed bibliography is fragmented by whitespace or line breaks. A longer candidate (10.1162/tacl\\_a\\_00595/2157340/tacl\\_a\\_00595.pdf) was visible in the surrounding text but could not be confirmed against doi.org as printed.","snippet":"Zhang, Xinyu and Thakur, Nandan and Ogundepo, Odunayo and Kamalloo, Ehsan and Alfonso-Hermelo, David and Li, Xiaoguang and Liu, Qun and Rezagholizadeh, Mehdi and Lin, Jimmy , doi =. Transactions of the Association for Computational Linguist","arxiv_id":"2605.22544","detector":"doi_compliance","evidence":{"ref_index":29,"verdict_class":"incontrovertible","resolved_title":null,"printed_excerpt":"Zhang, Xinyu and Thakur, Nandan and Ogundepo, Odunayo and Kamalloo, Ehsan and Alfonso-Hermelo, David and Li, Xiaoguang and Liu, Qun and Rezagholizadeh, Mehdi and Lin, Jimmy , doi =. Transactions of the Association for Computational Linguist","reconstructed_doi":"10.1162/tacl\\_a\\_00595/2157340/tacl\\_a\\_00595.pdf"},"severity":"advisory","ref_index":29,"audited_at":"2026-05-22T06:39:56.712357Z","event_type":"pith.integrity.v1","detected_doi":"10.1162/tacl\\_a\\_00595/2157340/tacl\\_a\\_00595.pdf","detector_url":"https://pith.science/pith-integrity-protocol#doi_compliance","external_url":null,"finding_type":"recoverable_identifier","evidence_hash":"c7f1b48841a8b351687730900ad8496918134181a5faead7eb5bc7953413336a","paper_version":1,"verdict_class":"incontrovertible","resolved_title":null,"detector_version":"1.0.0","detected_arxiv_id":null,"integrity_event_id":6266,"payload_sha256":"122b2220d0b0c08df89e7227aae0f44b16c6b9abd098119a6fef9a0ed66dcf0b","signature_b64":"KyTvy37lKOVBGEJSKgm+VcXkcDM1fAbilTnNg0oldPLPbBFUewQyA4vJN6vAxoRzC/he4tuLE/1nug8KzsfSDA==","signing_key_id":"pith-v1-2026-05"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-22T06:40:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"+AjBMH15xFY/+oUJ0rH72IFY/bJeWwoGuOTt0HAyNJdPHfEhl7TGha5QruPP+C5gcQFn+P65j9TbipjfxkmICw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T22:39:41.781846Z"},"content_sha256":"1fe4b254eaa6096e1e7f20efcfd3b5b5c71969599efc9082f2ea34dacd22321c","schema_version":"1.0","event_id":"sha256:1fe4b254eaa6096e1e7f20efcfd3b5b5c71969599efc9082f2ea34dacd22321c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q24F4WTDUUZJBKFTRIHYLH3REB/bundle.json","state_url":"https://pith.science/pith/Q24F4WTDUUZJBKFTRIHYLH3REB/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q24F4WTDUUZJBKFTRIHYLH3REB/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-05-26T22:39:41Z","links":{"resolver":"https://pith.science/pith/Q24F4WTDUUZJBKFTRIHYLH3REB","bundle":"https://pith.science/pith/Q24F4WTDUUZJBKFTRIHYLH3REB/bundle.json","state":"https://pith.science/pith/Q24F4WTDUUZJBKFTRIHYLH3REB/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q24F4WTDUUZJBKFTRIHYLH3REB/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2026:Q24F4WTDUUZJBKFTRIHYLH3REB","merge_version":"pith-open-graph-merge-v1","event_count":3,"valid_event_count":3,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"4e20f71ee5616c5cc507607d492961abe5750987a67f1c05c55ce506af380193","cross_cats_sorted":["cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-21T14:27:46Z","title_canon_sha256":"576af77fe98996cd3d6760a49ddcff94fdd18a1544ad9ff8b8792152ee644a68"},"schema_version":"1.0","source":{"id":"2605.22544","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2605.22544","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"arxiv_version","alias_value":"2605.22544v1","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2605.22544","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"pith_short_12","alias_value":"Q24F4WTDUUZJ","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"pith_short_16","alias_value":"Q24F4WTDUUZJBKFT","created_at":"2026-05-22T01:04:56Z"},{"alias_kind":"pith_short_8","alias_value":"Q24F4WTD","created_at":"2026-05-22T01:04:56Z"}],"graph_snapshots":[{"event_id":"sha256:441f15dfda18647f4bdf42504b51fa4957ae138680d36a88af6a90e91f808d6c","target":"graph","created_at":"2026-05-22T01:04:56Z","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/2605.22544/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Instruction embedding models have become common among state-of-the-art models, however are evaluated using a single prompt per task. The single-point evaluation ignores a main problem of the instruction-based approach namely: sensitivity to the phrasing of the instruction. We present an empirical study of prompt sensitivity across 6 embedding models, 11 datasets, and 15 task-specific prompts per dataset, a total of 990. We show that reported scores misrepresent the distribution of scores over plausible prompts. The default prompt can both systematically understate or overstate performance. Fur","authors_text":"Kenneth Enevoldsen, Yevhen Kostiuk","cross_cats":["cs.IR"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-21T14:27:46Z","title":"One prompt is not enough: Instruction Sensitivity Undermines Embedding Model Evaluation"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2605.22544","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:e84c5193f11d1e828dc7913f106b8e2cec8355dcd89d9fdc6a7350c19b285c4c","target":"record","created_at":"2026-05-22T01:04:56Z","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":"4e20f71ee5616c5cc507607d492961abe5750987a67f1c05c55ce506af380193","cross_cats_sorted":["cs.IR"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2026-05-21T14:27:46Z","title_canon_sha256":"576af77fe98996cd3d6760a49ddcff94fdd18a1544ad9ff8b8792152ee644a68"},"schema_version":"1.0","source":{"id":"2605.22544","kind":"arxiv","version":1}},"canonical_sha256":"86b85e5a63a53290a8b38a0f859f71207df16b4f2642e6762684fdad869f3fba","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"86b85e5a63a53290a8b38a0f859f71207df16b4f2642e6762684fdad869f3fba","first_computed_at":"2026-05-22T01:04:56.575939Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-22T01:04:56.575939Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"lb3b1tlK+gttWOE2wfiMBRANUFJtXfpBl1iNok68QmS5oCa1geaxRuHdmnUK57BTKfzz9uvPZkWUUe+Kd6W7BA==","signature_status":"signed_v1","signed_at":"2026-05-22T01:04:56.576770Z","signed_message":"canonical_sha256_bytes"},"source_id":"2605.22544","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:e84c5193f11d1e828dc7913f106b8e2cec8355dcd89d9fdc6a7350c19b285c4c","sha256:441f15dfda18647f4bdf42504b51fa4957ae138680d36a88af6a90e91f808d6c","sha256:1fe4b254eaa6096e1e7f20efcfd3b5b5c71969599efc9082f2ea34dacd22321c"],"state_sha256":"a54301b9e39a0f156c92ddb8dc699443b783b7075608220efb8d3dfda2ac2771"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"OXuhDIwTFUhnBWYRmodHpotgI6bFupuSnFvaGBHFj4j+NCRTDVhdZ0NBenxwZVnyWyFD2VYIzRfDf+FMRh5PCA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T22:39:41.784910Z","bundle_sha256":"e26fa53a7ecbc6943c26e1c959e482c94b75f784da57320c2452ca1f19f4a58d"}}