{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:X4WHQBUPY37BMSE2SEJCOCR36R","short_pith_number":"pith:X4WHQBUP","canonical_record":{"source":{"id":"2405.01588","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-29T02:26:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4da5c117af5fa9c5368a605145472af8d457bfdc2c0f47ec5cdfcb70c7a2a606","abstract_canon_sha256":"dd685a44b328952dcdd0ec752d94f2b9e52018c681ffff2b36a42ef6ff0b317e"},"schema_version":"1.0"},"canonical_sha256":"bf2c78068fc6fe16489a9112270a3bf458d96e36f9df59b62be7336bed2a8b90","source":{"kind":"arxiv","id":"2405.01588","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.01588","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"arxiv_version","alias_value":"2405.01588v1","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.01588","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"pith_short_12","alias_value":"X4WHQBUPY37B","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"pith_short_16","alias_value":"X4WHQBUPY37BMSE2","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"pith_short_8","alias_value":"X4WHQBUP","created_at":"2026-07-05T08:14:58Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:X4WHQBUPY37BMSE2SEJCOCR36R","target":"record","payload":{"canonical_record":{"source":{"id":"2405.01588","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-29T02:26:15Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"4da5c117af5fa9c5368a605145472af8d457bfdc2c0f47ec5cdfcb70c7a2a606","abstract_canon_sha256":"dd685a44b328952dcdd0ec752d94f2b9e52018c681ffff2b36a42ef6ff0b317e"},"schema_version":"1.0"},"canonical_sha256":"bf2c78068fc6fe16489a9112270a3bf458d96e36f9df59b62be7336bed2a8b90","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:14:58.549401Z","signature_b64":"uj1Fu/OUZKkDnXQTwPVEgZrrpgMeqPgeSPINHy143VT8DhTPoQOvmqfSaGT9m4NKEFr4THAvIqYv11hIEVJsBg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"bf2c78068fc6fe16489a9112270a3bf458d96e36f9df59b62be7336bed2a8b90","last_reissued_at":"2026-07-05T08:14:58.549006Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:14:58.549006Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2405.01588","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":"2026-07-05T08:14:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"3PHMcBKF2lMwGt11iB4S40IN7BSEUWFmnXtM2dl+0peDihcRZ7cQ9C4FOdFKjXWdKl3nqWOlo3EnTdNPi6BYBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-10T14:53:46.995615Z"},"content_sha256":"d3945a188b257af8d9de456b89a0627cd2bebc33df381bce8fc7c7f0e331f0a6","schema_version":"1.0","event_id":"sha256:d3945a188b257af8d9de456b89a0627cd2bebc33df381bce8fc7c7f0e331f0a6"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:X4WHQBUPY37BMSE2SEJCOCR36R","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Towards Unbiased Evaluation of Detecting Unanswerable Questions in EHRSQL","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.CL","authors_text":"Edward Choi, Gyubok Lee, Sangmook Kim, Se-Young Yun, Sihyeon Kim, Yongjin Yang","submitted_at":"2024-04-29T02:26:15Z","abstract_excerpt":"Incorporating unanswerable questions into EHR QA systems is crucial for testing the trustworthiness of a system, as providing non-existent responses can mislead doctors in their diagnoses. The EHRSQL dataset stands out as a promising benchmark because it is the only dataset that incorporates unanswerable questions in the EHR QA system alongside practical questions. However, in this work, we identify a data bias in these unanswerable questions; they can often be discerned simply by filtering with specific N-gram patterns. Such biases jeopardize the authenticity and reliability of QA system eval"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.01588","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/2405.01588/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":"2026-07-05T08:14:58Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"wgFfDGuvLxBic7cS2vjGBbzHIpUOAlKTq8zbwOVA4WkMnQOeiMISSz+YcAP05Og4NHkhtvNkBM8cA8vmcoV3Bw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-10T14:53:46.995980Z"},"content_sha256":"8780025a81dffcdeca268dab5bf34880f6b4cea09827054636401224dc89dc8c","schema_version":"1.0","event_id":"sha256:8780025a81dffcdeca268dab5bf34880f6b4cea09827054636401224dc89dc8c"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/X4WHQBUPY37BMSE2SEJCOCR36R/bundle.json","state_url":"https://pith.science/pith/X4WHQBUPY37BMSE2SEJCOCR36R/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/X4WHQBUPY37BMSE2SEJCOCR36R/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":"2026-07-10T14:53:46Z","links":{"resolver":"https://pith.science/pith/X4WHQBUPY37BMSE2SEJCOCR36R","bundle":"https://pith.science/pith/X4WHQBUPY37BMSE2SEJCOCR36R/bundle.json","state":"https://pith.science/pith/X4WHQBUPY37BMSE2SEJCOCR36R/state.json","well_known_bundle":"https://pith.science/.well-known/pith/X4WHQBUPY37BMSE2SEJCOCR36R/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:X4WHQBUPY37BMSE2SEJCOCR36R","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":"dd685a44b328952dcdd0ec752d94f2b9e52018c681ffff2b36a42ef6ff0b317e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-29T02:26:15Z","title_canon_sha256":"4da5c117af5fa9c5368a605145472af8d457bfdc2c0f47ec5cdfcb70c7a2a606"},"schema_version":"1.0","source":{"id":"2405.01588","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2405.01588","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"arxiv_version","alias_value":"2405.01588v1","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2405.01588","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"pith_short_12","alias_value":"X4WHQBUPY37B","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"pith_short_16","alias_value":"X4WHQBUPY37BMSE2","created_at":"2026-07-05T08:14:58Z"},{"alias_kind":"pith_short_8","alias_value":"X4WHQBUP","created_at":"2026-07-05T08:14:58Z"}],"graph_snapshots":[{"event_id":"sha256:8780025a81dffcdeca268dab5bf34880f6b4cea09827054636401224dc89dc8c","target":"graph","created_at":"2026-07-05T08:14:58Z","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/2405.01588/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Incorporating unanswerable questions into EHR QA systems is crucial for testing the trustworthiness of a system, as providing non-existent responses can mislead doctors in their diagnoses. The EHRSQL dataset stands out as a promising benchmark because it is the only dataset that incorporates unanswerable questions in the EHR QA system alongside practical questions. However, in this work, we identify a data bias in these unanswerable questions; they can often be discerned simply by filtering with specific N-gram patterns. Such biases jeopardize the authenticity and reliability of QA system eval","authors_text":"Edward Choi, Gyubok Lee, Sangmook Kim, Se-Young Yun, Sihyeon Kim, Yongjin Yang","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-29T02:26:15Z","title":"Towards Unbiased Evaluation of Detecting Unanswerable Questions in EHRSQL"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2405.01588","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:d3945a188b257af8d9de456b89a0627cd2bebc33df381bce8fc7c7f0e331f0a6","target":"record","created_at":"2026-07-05T08:14:58Z","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":"dd685a44b328952dcdd0ec752d94f2b9e52018c681ffff2b36a42ef6ff0b317e","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.CL","submitted_at":"2024-04-29T02:26:15Z","title_canon_sha256":"4da5c117af5fa9c5368a605145472af8d457bfdc2c0f47ec5cdfcb70c7a2a606"},"schema_version":"1.0","source":{"id":"2405.01588","kind":"arxiv","version":1}},"canonical_sha256":"bf2c78068fc6fe16489a9112270a3bf458d96e36f9df59b62be7336bed2a8b90","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"bf2c78068fc6fe16489a9112270a3bf458d96e36f9df59b62be7336bed2a8b90","first_computed_at":"2026-07-05T08:14:58.549006Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:14:58.549006Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"uj1Fu/OUZKkDnXQTwPVEgZrrpgMeqPgeSPINHy143VT8DhTPoQOvmqfSaGT9m4NKEFr4THAvIqYv11hIEVJsBg==","signature_status":"signed_v1","signed_at":"2026-07-05T08:14:58.549401Z","signed_message":"canonical_sha256_bytes"},"source_id":"2405.01588","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d3945a188b257af8d9de456b89a0627cd2bebc33df381bce8fc7c7f0e331f0a6","sha256:8780025a81dffcdeca268dab5bf34880f6b4cea09827054636401224dc89dc8c"],"state_sha256":"6bb1e7dd3e0e54dc2e3655328c37e38b384ba3b1b79dda422e4a8e48d5c996de"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6xR7pJAjbWmuY7Uav/x0Xslto87KZCJMP1JmKGd4IJZVcYq1bFUpuGlhXzbD/rlsK8CiBHIDN4SF+37j4h9xDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-10T14:53:46.997943Z","bundle_sha256":"fae8d984747677dc041e7bd89c93b3a3045903c8803ee91a8f3a8a45907d7316"}}