{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2024:DEEZKVVDZEH67TNAD5H53HQJCK","short_pith_number":"pith:DEEZKVVD","canonical_record":{"source":{"id":"2407.07313","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T02:20:19Z","cross_cats_sorted":[],"title_canon_sha256":"a425603c11e711217a049b7df48be38bbdce7e2dd6569ce461938e09d2a3de7e","abstract_canon_sha256":"c84d4337de93c2c69ffd18f8dbc07d6990a5edf71c1165f43ed63dcdb0d7cb55"},"schema_version":"1.0"},"canonical_sha256":"19099556a3c90fefcda01f4fdd9e09128ce8cdb330f260f9388655a30156919e","source":{"kind":"arxiv","id":"2407.07313","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.07313","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"arxiv_version","alias_value":"2407.07313v4","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.07313","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"pith_short_12","alias_value":"DEEZKVVDZEH6","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"pith_short_16","alias_value":"DEEZKVVDZEH67TNA","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"pith_short_8","alias_value":"DEEZKVVD","created_at":"2026-07-05T11:22:31Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2024:DEEZKVVDZEH67TNAD5H53HQJCK","target":"record","payload":{"canonical_record":{"source":{"id":"2407.07313","kind":"arxiv","version":4},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T02:20:19Z","cross_cats_sorted":[],"title_canon_sha256":"a425603c11e711217a049b7df48be38bbdce7e2dd6569ce461938e09d2a3de7e","abstract_canon_sha256":"c84d4337de93c2c69ffd18f8dbc07d6990a5edf71c1165f43ed63dcdb0d7cb55"},"schema_version":"1.0"},"canonical_sha256":"19099556a3c90fefcda01f4fdd9e09128ce8cdb330f260f9388655a30156919e","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:22:31.306056Z","signature_b64":"I3LRlCPjX+LoYZAVmzCNTC9msG28LMfSNVNRruIWy7EWFZ2vF/uCpxgwGyTiG8tQ3eq376dCCSeQegKeKjVsAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"19099556a3c90fefcda01f4fdd9e09128ce8cdb330f260f9388655a30156919e","last_reissued_at":"2026-07-05T11:22:31.305582Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:22:31.305582Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2407.07313","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T11:22:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k8FCFL5VtxI6fM5xSobcfdMdhXN02ZQUeolrPtk8MJvOBB8k/wjvenbCNb0wfT2P4EgI1R44y1v+RqynTOWnBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T06:13:49.990441Z"},"content_sha256":"d5c76d1b607f41c52a3c2c970944a1985b586b8092abff06d587f0d9cd06e944","schema_version":"1.0","event_id":"sha256:d5c76d1b607f41c52a3c2c970944a1985b586b8092abff06d587f0d9cd06e944"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2024:DEEZKVVDZEH67TNAD5H53HQJCK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"ETM: Modern Insights into Perspective on Text-to-SQL Evaluation in the Age of Large Language Models","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":[],"primary_cat":"cs.CL","authors_text":"Benjamin G. Ascoli, Jinho D. Choi, Yasoda Sai Ram Kandikonda","submitted_at":"2024-07-10T02:20:19Z","abstract_excerpt":"The task of Text-to-SQL enables anyone to retrieve information from SQL databases using natural language. While this task has made substantial progress, the two primary evaluation metrics - Execution Accuracy (EXE) and Exact Set Matching Accuracy (ESM) - suffer from inherent limitations that can misrepresent performance. Specifically, ESM's rigid matching overlooks semantically correct but stylistically different queries, whereas EXE can overestimate correctness by ignoring structural errors that yield correct outputs. These shortcomings become especially problematic when assessing outputs fro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.07313","kind":"arxiv","version":4},"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/2407.07313/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":"2026-07-05T11:22:31Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"t9+7FGSO1JR4iGQ8Z4tbRWdnOUfBCv/wSeqap8+73PqpTSQQl/hujrYri0C7npCjeAWoE/ymWygUAOhPqxeGBA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-09T06:13:49.990971Z"},"content_sha256":"7368c6078b0acf986378705fe402e7cc148a5710e22b146f6c5d89d8af3a14ab","schema_version":"1.0","event_id":"sha256:7368c6078b0acf986378705fe402e7cc148a5710e22b146f6c5d89d8af3a14ab"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/DEEZKVVDZEH67TNAD5H53HQJCK/bundle.json","state_url":"https://pith.science/pith/DEEZKVVDZEH67TNAD5H53HQJCK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/DEEZKVVDZEH67TNAD5H53HQJCK/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":"2026-08-09T06:13:49Z","links":{"resolver":"https://pith.science/pith/DEEZKVVDZEH67TNAD5H53HQJCK","bundle":"https://pith.science/pith/DEEZKVVDZEH67TNAD5H53HQJCK/bundle.json","state":"https://pith.science/pith/DEEZKVVDZEH67TNAD5H53HQJCK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/DEEZKVVDZEH67TNAD5H53HQJCK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2024:DEEZKVVDZEH67TNAD5H53HQJCK","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":"c84d4337de93c2c69ffd18f8dbc07d6990a5edf71c1165f43ed63dcdb0d7cb55","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T02:20:19Z","title_canon_sha256":"a425603c11e711217a049b7df48be38bbdce7e2dd6569ce461938e09d2a3de7e"},"schema_version":"1.0","source":{"id":"2407.07313","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2407.07313","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"arxiv_version","alias_value":"2407.07313v4","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2407.07313","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"pith_short_12","alias_value":"DEEZKVVDZEH6","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"pith_short_16","alias_value":"DEEZKVVDZEH67TNA","created_at":"2026-07-05T11:22:31Z"},{"alias_kind":"pith_short_8","alias_value":"DEEZKVVD","created_at":"2026-07-05T11:22:31Z"}],"graph_snapshots":[{"event_id":"sha256:7368c6078b0acf986378705fe402e7cc148a5710e22b146f6c5d89d8af3a14ab","target":"graph","created_at":"2026-07-05T11:22:31Z","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/2407.07313/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The task of Text-to-SQL enables anyone to retrieve information from SQL databases using natural language. While this task has made substantial progress, the two primary evaluation metrics - Execution Accuracy (EXE) and Exact Set Matching Accuracy (ESM) - suffer from inherent limitations that can misrepresent performance. Specifically, ESM's rigid matching overlooks semantically correct but stylistically different queries, whereas EXE can overestimate correctness by ignoring structural errors that yield correct outputs. These shortcomings become especially problematic when assessing outputs fro","authors_text":"Benjamin G. Ascoli, Jinho D. Choi, Yasoda Sai Ram Kandikonda","cross_cats":[],"headline":"","license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T02:20:19Z","title":"ETM: Modern Insights into Perspective on Text-to-SQL Evaluation in the Age of Large Language Models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2407.07313","kind":"arxiv","version":4},"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:d5c76d1b607f41c52a3c2c970944a1985b586b8092abff06d587f0d9cd06e944","target":"record","created_at":"2026-07-05T11:22:31Z","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":"c84d4337de93c2c69ffd18f8dbc07d6990a5edf71c1165f43ed63dcdb0d7cb55","cross_cats_sorted":[],"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CL","submitted_at":"2024-07-10T02:20:19Z","title_canon_sha256":"a425603c11e711217a049b7df48be38bbdce7e2dd6569ce461938e09d2a3de7e"},"schema_version":"1.0","source":{"id":"2407.07313","kind":"arxiv","version":4}},"canonical_sha256":"19099556a3c90fefcda01f4fdd9e09128ce8cdb330f260f9388655a30156919e","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"19099556a3c90fefcda01f4fdd9e09128ce8cdb330f260f9388655a30156919e","first_computed_at":"2026-07-05T11:22:31.305582Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T11:22:31.305582Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"I3LRlCPjX+LoYZAVmzCNTC9msG28LMfSNVNRruIWy7EWFZ2vF/uCpxgwGyTiG8tQ3eq376dCCSeQegKeKjVsAw==","signature_status":"signed_v1","signed_at":"2026-07-05T11:22:31.306056Z","signed_message":"canonical_sha256_bytes"},"source_id":"2407.07313","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:d5c76d1b607f41c52a3c2c970944a1985b586b8092abff06d587f0d9cd06e944","sha256:7368c6078b0acf986378705fe402e7cc148a5710e22b146f6c5d89d8af3a14ab"],"state_sha256":"05d0ed6c08779aef43d0d8fbf9bcb16bc41a221b9bc7a4e9485c8e5107f07131"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"CxtjhjEZVIzoXbHWCA6/oMyauHQ1cf7Vs+HSj2kDusukySi95oCoNu7S/xeqUM1qsepHJ1JLJybeiYyG8OHNBg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-09T06:13:49.995888Z","bundle_sha256":"91ee1069ab577272ac48604b22a39a886ba30c9f092232259bd2f2b7b1d5ac74"}}