{"observation":{"id":"df4d4c76-2422-413b-9d07-a3489b53cb50","tool":"datalab","tool_name":"Datalab","criterion":"structural-clean-output","criterion_name":"Structural Clean Output","criterion_definition":"Is the JSON directly consumable by a downstream AI pipeline or system without requiring a structural transformation layer?","criterion_evidence_type":"transformation","criterion_rank_role":"context","criterion_rank_role_reason":"Directly consumable JSON is valuable for workflow convenience, but it is a delivery/integration concern rather than the core measure of extraction quality itself. (3 of 3 judges)","scenario":"invoice-pdf","scenario_name":"Invoice PDF","group_tag":"financial-document-extraction","scenario_description":"A 2-page broadcast advertising invoice PDF with 8 line items, complex time/day fields, large dollar amounts, and compliance text, used to test hierarchical line-item extraction and financial validation.","modality":"pdf","input_text":null,"input_artifact_refs":[{"alt":null,"url":null,"role":null,"filename":"Invoice PDF.pdf"}],"stresses":["Nested line-item hierarchy extraction","Multi-page line-item continuity across a page break","Large dollar amount precision and total validation","Parsing time slots, day patterns, and air dates","Extraction of alphanumeric ad IDs and reference codes","Structured metadata mapping for advertiser, station, billing, and remit sections","Political advertising and FCC compliance text recognition"],"verdict":"mixed","score":null,"score_total":null,"note":"The generated JSON is structurally valid but not order-stable: line_items appears before other top-level sections that were defined earlier in the supplied schema.","evidence_state":"verified","source":null,"artifacts":[{"url":"https://d3epheqghktydj.cloudfront.net/research-media-invoice-schema-order-1938d10e3382.png","role":"context","alt":null},{"url":"https://d3epheqghktydj.cloudfront.net/research-media-datalab-invoice-schema-order-53439296a014.png","role":"output","alt":null}],"run_id":"ec4d736d-95f9-4c88-884c-e280435f7b7b","study_title":"Extract and query structured data from documents using natural language","study_kind":"generation","research_task":"86b9y25e5","tested_at":null,"completeness":"output-only","input":{"state":"not-captured","text":null,"files":[],"modality":"pdf","stresses":["Nested line-item hierarchy extraction","Multi-page line-item continuity across a page break","Large dollar amount precision and total validation","Parsing time slots, day patterns, and air dates","Extraction of alphanumeric ad IDs and reference codes","Structured metadata mapping for advertiser, station, billing, and remit sections","Political advertising and FCC compliance text recognition"]},"tool_page_slug":"datalab","tool_url":"https://aidemos.com/tools/datalab","permalink":"https://aidemos.com/evidence/df4d4c76-2422-413b-9d07-a3489b53cb50","api_url":"https://ai.aidemos.com/v1/observations/df4d4c76-2422-413b-9d07-a3489b53cb50"},"peers":[{"id":"be4e0664-3a37-4913-a8df-3cd6f8915087","tool":"extend-ai","tool_name":"Extend AI","verdict":"failed","score":null,"score_total":null,"note":"It reorders several schema sections relative to the supplied schema, so the output is not consistently in the authored field order.","artifact_count":3,"thumbnail":"https://cdn.futuresmart.ai/public/aidemos/9e7bb7c55fbf41d8bf08b8139a192ae4.png?v=1","evidence_url":"https://aidemos.com/evidence/be4e0664-3a37-4913-a8df-3cd6f8915087"},{"id":"b78d37af-33a7-4f25-91f5-6a7c39e2f334","tool":"retab","tool_name":"Retab","verdict":"mixed","score":null,"score_total":null,"note":"Does not preserve the schema-defined property order, so consumers that rely on key order need an extra formatting pass.","artifact_count":2,"thumbnail":"https://d3epheqghktydj.cloudfront.net/research-media-invoice-schema-order-85baff3c493d.png","evidence_url":"https://aidemos.com/evidence/b78d37af-33a7-4f25-91f5-6a7c39e2f334"}],"other_criteria":[{"id":"76f5a5f0-1389-4341-a623-595405cf206b","criterion":"extraction-accuracy","criterion_name":"Extraction Accuracy","rank_role":"decisive","verdict":"worked","score":null,"score_total":null,"note":"Extracts the invoice metadata and monetary values accurately, including invoice 4064621-1, invoice date 10/28/12, gross_total 29750, agency_commission 4462.5, net_amount_due 25287.5, and payment_terms 30 Days.","artifact_count":2,"evidence_url":"https://aidemos.com/evidence/76f5a5f0-1389-4341-a623-595405cf206b"},{"id":"c01474f8-d601-4535-b19c-9ab5759d7f35","criterion":"schema-adherence","criterion_name":"Schema Adherence","rank_role":"decisive","verdict":"worked","score":null,"score_total":null,"note":"Reconstructs the invoice into nested JSON objects and arrays that follow the supplied schema, including invoice_metadata and line_items rather than returning generic OCR text.","artifact_count":5,"evidence_url":"https://aidemos.com/evidence/c01474f8-d601-4535-b19c-9ab5759d7f35"},{"id":"f376333b-6c46-4298-91ee-eddd286afd5d","criterion":"semantic-field-enrichment","criterion_name":"Semantic Field Enrichment","rank_role":"decisive","verdict":"worked","score":null,"score_total":null,"note":"Correctly extracts derived scheduling fields on line 2, including day_of_week Su, time_slot 9a-10a, air_time 9:38 AM, and ad_id NRCCWI071005.","artifact_count":1,"evidence_url":"https://aidemos.com/evidence/f376333b-6c46-4298-91ee-eddd286afd5d"},{"id":"a2dfbd36-ebfc-4d14-ab59-0d615bee1ed4","criterion":"table-and-record-completeness","criterion_name":"Table & Record Completeness","rank_role":"decisive","verdict":"worked","score":null,"score_total":null,"note":"Keeps all 8 advertising spots as separate line-item records, with the JSON tree showing indices 0 through 7 and no adjacent-row merging.","artifact_count":2,"evidence_url":"https://aidemos.com/evidence/a2dfbd36-ebfc-4d14-ab59-0d615bee1ed4"}],"appears_in":[{"page_type":"ranking","slug":"document-extraction","title":"Best AI Tools for Extracting Structured Data from PDFs and Business Documents","url":"https://aidemos.com/best/document-extraction","binding":"run"},{"page_type":"tool","slug":"unstract","title":null,"url":"https://aidemos.com/tools/unstract","binding":"run"}],"same_scenario":[{"id":"be4e0664-3a37-4913-a8df-3cd6f8915087","tool":"extend-ai","tool_name":"Extend AI","verdict":"failed","score":null,"score_total":null,"note":"It reorders several schema sections relative to the supplied schema, so the output is not consistently in the authored field order."},{"id":"b78d37af-33a7-4f25-91f5-6a7c39e2f334","tool":"retab","tool_name":"Retab","verdict":"mixed","score":null,"score_total":null,"note":"Does not preserve the schema-defined property order, so consumers that rely on key order need an extra formatting pass."}]}