| <!doctype html> |
| <html lang="en"> |
| <head> |
| <meta charset="utf-8"> |
| <meta name="viewport" content="width=device-width, initial-scale=1"> |
| <base target="_blank"> |
| <title>LongListBench Leaderboard</title> |
| <style> |
| @font-face { |
| font-family: "Suisse Intl"; |
| src: local("Suisse Intl"), local("SuisseIntl-Regular"); |
| font-style: normal; |
| font-weight: 400; |
| font-display: swap; |
| } |
| @font-face { |
| font-family: "Suisse Intl"; |
| src: local("Suisse Intl Medium"), local("SuisseIntl-Medium"); |
| font-style: normal; |
| font-weight: 600; |
| font-display: swap; |
| } |
| :root { |
| --paper: #f9f7f3; |
| --surface: #efede8; |
| --surface-strong: #e8e5df; |
| --ink: #363636; |
| --muted: #77756f; |
| --line: #d7d4cd; |
| --signal: #ffe100; |
| --signal-soft: #fff6ae; |
| --black: #1d1d1b; |
| --white: #ffffff; |
| } |
| * { box-sizing: border-box; } |
| html { scroll-behavior: smooth; } |
| body { |
| margin: 0; |
| background: var(--paper); |
| color: var(--ink); |
| font-family: "Suisse Intl", "Helvetica Neue", Arial, sans-serif; |
| font-size: 16px; |
| line-height: 1.6; |
| -webkit-font-smoothing: antialiased; |
| text-rendering: optimizeLegibility; |
| } |
| a { color: inherit; } |
| a:focus-visible { outline: 3px solid var(--signal); outline-offset: 3px; } |
| .shell { |
| width: min(1280px, calc(100% - 32px)); |
| margin: 0 auto; |
| } |
| .leaderboard-section { padding-top: 12px; } |
| .tablebox { |
| overflow-x: auto; |
| border: 1px solid var(--line); |
| border-radius: 20px; |
| background: rgba(255,255,255,.62); |
| } |
| table { width: 100%; min-width: 1020px; border-collapse: collapse; font-size: 14px; } |
| th, td { |
| padding: 18px 17px; |
| border-top: 1px solid #e4e1db; |
| text-align: left; |
| white-space: nowrap; |
| } |
| thead th { |
| border-top: 0; |
| color: var(--muted); |
| font: 10px/1.2 ui-monospace, SFMono-Regular, Menlo, monospace; |
| letter-spacing: .055em; |
| text-transform: uppercase; |
| } |
| .sort-button { |
| padding: 0; |
| display: inline-flex; |
| align-items: center; |
| gap: 6px; |
| border: 0; |
| background: transparent; |
| color: inherit; |
| cursor: pointer; |
| font: inherit; |
| letter-spacing: inherit; |
| text-transform: inherit; |
| white-space: nowrap; |
| } |
| .column-actions { |
| display: flex; |
| align-items: center; |
| gap: 5px; |
| } |
| th.numeric .column-actions { justify-content: flex-end; } |
| .sort-button:hover, .sort-button:focus-visible { color: var(--ink); } |
| .sort-button:focus-visible { |
| border-radius: 3px; |
| outline: 3px solid var(--signal); |
| outline-offset: 3px; |
| } |
| .sort-arrow { |
| width: 11px; |
| color: #aaa79f; |
| font-size: 11px; |
| text-align: center; |
| } |
| th[aria-sort="ascending"] .sort-button, |
| th[aria-sort="descending"] .sort-button { color: var(--ink); } |
| th[aria-sort="ascending"] .sort-arrow, |
| th[aria-sort="descending"] .sort-arrow { color: var(--ink); } |
| .definition-button { |
| width: 16px; |
| height: 16px; |
| padding: 0; |
| display: inline-grid; |
| place-items: center; |
| flex: 0 0 auto; |
| border: 1px solid #c5c2bb; |
| border-radius: 50%; |
| background: rgba(255,255,255,.5); |
| color: var(--muted); |
| cursor: pointer; |
| font: 9px/1 ui-monospace, SFMono-Regular, Menlo, monospace; |
| text-transform: lowercase; |
| } |
| .definition-button:hover, |
| .definition-button[aria-expanded="true"] { |
| border-color: var(--ink); |
| background: var(--ink); |
| color: var(--white); |
| } |
| .definition-button:focus-visible { |
| outline: 3px solid var(--signal); |
| outline-offset: 2px; |
| } |
| tbody > tr[data-result-row] { transition: background-color .16s ease; } |
| tbody > tr[data-result-row]:hover { background: rgba(255,255,255,.68); } |
| td.metric-best { |
| background: rgba(255,225,0,.16); |
| font-weight: 600; |
| } |
| tbody > tr[data-result-row]:hover > td.metric-best { |
| background: rgba(255,225,0,.24); |
| } |
| .configuration { min-width: 270px; } |
| .model-line { display: flex; align-items: center; gap: 8px; } |
| .model-line strong { font-size: 15px; font-weight: 600; } |
| .configuration > span:not(.protocol) { |
| display: block; |
| margin-top: 2px; |
| color: var(--muted); |
| font-size: 11px; |
| } |
| .configuration-actions { |
| margin-top: 7px; |
| display: flex; |
| align-items: center; |
| gap: 7px; |
| } |
| .protocol { |
| display: inline-block; |
| padding: 3px 6px; |
| border: 1px solid var(--line); |
| border-radius: 4px; |
| color: var(--muted); |
| font: 9px/1 ui-monospace, SFMono-Regular, Menlo, monospace; |
| letter-spacing: .04em; |
| text-transform: uppercase; |
| } |
| .details-button { |
| padding: 3px 7px; |
| display: inline-flex; |
| align-items: center; |
| gap: 6px; |
| border: 1px solid #bbb8b0; |
| border-radius: 4px; |
| background: rgba(255,255,255,.56); |
| color: var(--ink); |
| cursor: pointer; |
| font: 9px/1 ui-monospace, SFMono-Regular, Menlo, monospace; |
| letter-spacing: .04em; |
| text-transform: uppercase; |
| } |
| .details-button:hover, |
| .details-button[aria-expanded="true"] { |
| border-color: var(--ink); |
| background: var(--ink); |
| color: var(--white); |
| } |
| .details-button:focus-visible { |
| outline: 3px solid var(--signal); |
| outline-offset: 2px; |
| } |
| .details-symbol { width: 8px; text-align: center; } |
| .detail-row[hidden] { display: none; } |
| .detail-row > td { |
| padding: 0; |
| border-top: 0; |
| white-space: normal; |
| } |
| .detail-panel { |
| padding: 14px 16px 16px; |
| border-top: 1px solid var(--line); |
| border-bottom: 1px solid var(--line); |
| background: var(--surface-strong); |
| } |
| .detail-panel-heading { |
| margin-bottom: 10px; |
| display: flex; |
| align-items: baseline; |
| justify-content: space-between; |
| gap: 16px; |
| } |
| .detail-panel-heading strong { font-size: 13px; font-weight: 600; } |
| .detail-panel-heading span { |
| color: var(--muted); |
| font: 9px/1 ui-monospace, SFMono-Regular, Menlo, monospace; |
| letter-spacing: .04em; |
| text-transform: uppercase; |
| } |
| .result-note { |
| margin-bottom: 10px; |
| padding: 10px 12px; |
| border: 1px solid var(--line); |
| border-radius: 8px; |
| background: rgba(255,255,255,.52); |
| color: var(--muted); |
| font-size: 11px; |
| line-height: 1.5; |
| } |
| .document-scroll { |
| max-height: 380px; |
| overflow: auto; |
| border: 1px solid #cfccc4; |
| border-radius: 10px; |
| background: rgba(255,255,255,.72); |
| } |
| .document-table { |
| width: 100%; |
| min-width: 720px; |
| border-collapse: collapse; |
| font-size: 12px; |
| } |
| .document-table th, |
| .document-table td { |
| padding: 9px 11px; |
| border-top: 1px solid #e3e0d9; |
| white-space: nowrap; |
| } |
| .document-table thead th { |
| position: sticky; |
| top: 0; |
| z-index: 1; |
| border-top: 0; |
| background: #f1efea; |
| } |
| .document-table tbody tr:hover { background: rgba(255,255,255,.8); } |
| .document-name { |
| font: 10px/1.25 ui-monospace, SFMono-Regular, Menlo, monospace; |
| letter-spacing: .01em; |
| } |
| .document-status { |
| display: inline-flex; |
| align-items: center; |
| gap: 6px; |
| color: var(--muted); |
| font: 9px/1 ui-monospace, SFMono-Regular, Menlo, monospace; |
| letter-spacing: .03em; |
| text-transform: uppercase; |
| } |
| .document-status::before { |
| content: ""; |
| width: 6px; |
| height: 6px; |
| border-radius: 50%; |
| background: #b8b5ae; |
| } |
| .document-status.complete { color: var(--ink); } |
| .document-status.complete::before { background: var(--signal); } |
| .prompt-panel { |
| margin-top: 14px; |
| padding-top: 14px; |
| border-top: 1px solid var(--line); |
| } |
| .prompt-panel-heading { |
| margin-bottom: 10px; |
| display: flex; |
| align-items: baseline; |
| justify-content: space-between; |
| gap: 16px; |
| } |
| .prompt-panel-heading strong, |
| .prompt-template > strong { |
| font-size: 12px; |
| font-weight: 600; |
| } |
| .prompt-panel-heading span { |
| color: var(--muted); |
| font-size: 11px; |
| } |
| .prompt-template + .prompt-template { margin-top: 12px; } |
| .prompt-scroll { |
| max-height: 320px; |
| margin-top: 6px; |
| overflow: auto; |
| border: 1px solid #cfccc4; |
| border-radius: 10px; |
| background: rgba(255,255,255,.72); |
| } |
| .prompt-scroll pre { |
| margin: 0; |
| padding: 13px 14px; |
| white-space: pre-wrap; |
| overflow-wrap: anywhere; |
| font: 11px/1.55 ui-monospace, SFMono-Regular, Menlo, monospace; |
| } |
| .numeric { text-align: right; font-variant-numeric: tabular-nums; } |
| .price { font: 12px/1.2 ui-monospace, SFMono-Regular, Menlo, monospace; } |
| .cost-cell { |
| display: inline-flex; |
| align-items: center; |
| justify-content: flex-end; |
| gap: 6px; |
| } |
| .metric-cell { |
| display: inline-flex; |
| align-items: center; |
| justify-content: flex-end; |
| gap: 5px; |
| } |
| .beta-badge { |
| padding: 2px 4px; |
| border-radius: 3px; |
| background: var(--signal); |
| color: var(--black); |
| font: 8px/1 ui-monospace, SFMono-Regular, Menlo, monospace; |
| letter-spacing: .04em; |
| text-transform: uppercase; |
| } |
| .pricing-comparison { |
| color: rgba(255,255,255,.74); |
| font: 10px/1.5 ui-monospace, SFMono-Regular, Menlo, monospace; |
| } |
| .metric-popover { |
| position: fixed; |
| z-index: 50; |
| width: min(310px, calc(100vw - 24px)); |
| padding: 15px 16px 16px; |
| border-radius: 12px; |
| color: rgba(255,255,255,.74); |
| background: var(--black); |
| box-shadow: 0 16px 50px rgba(29,29,27,.24); |
| } |
| .metric-popover[hidden] { display: none; } |
| .metric-popover-label { |
| display: flex; |
| align-items: center; |
| gap: 7px; |
| color: rgba(255,255,255,.5); |
| font: 9px/1.2 ui-monospace, SFMono-Regular, Menlo, monospace; |
| letter-spacing: .06em; |
| text-transform: uppercase; |
| } |
| .metric-popover-label::before { |
| content: ""; |
| width: 6px; |
| height: 6px; |
| border-radius: 50%; |
| background: var(--signal); |
| } |
| .metric-popover strong { |
| display: block; |
| margin-top: 10px; |
| color: var(--white); |
| font-size: 15px; |
| font-weight: 600; |
| } |
| .metric-popover-copy { |
| margin: 5px 0 0; |
| } |
| .metric-popover-copy > p { |
| margin: 0; |
| font-size: 12px; |
| line-height: 1.5; |
| } |
| .pricing-comparison-table { |
| width: 100%; |
| min-width: 0; |
| margin-top: 10px; |
| border: 1px solid rgba(255,255,255,.18); |
| border-radius: 6px; |
| color: rgba(255,255,255,.74); |
| font-size: 10px; |
| } |
| .pricing-comparison-table th, |
| .pricing-comparison-table td { |
| padding: 7px 8px; |
| border-top: 1px solid rgba(255,255,255,.14); |
| } |
| .pricing-comparison-table thead th { |
| border-top: 0; |
| color: rgba(255,255,255,.5); |
| background: transparent; |
| font: 8px/1.2 ui-monospace, SFMono-Regular, Menlo, monospace; |
| } |
| .pricing-comparison-table tbody th { font-weight: 400; } |
| .pricing-comparison-table tbody tr.experimental th, |
| .pricing-comparison-table tbody tr.experimental td { |
| background: var(--signal); |
| color: var(--black); |
| font-weight: 600; |
| } |
| .submit-panel { |
| margin: 56px 0 20px; |
| padding: 24px 26px; |
| display: grid; |
| grid-template-columns: 1fr auto; |
| gap: 32px; |
| align-items: center; |
| border-radius: 22px; |
| background: var(--surface-strong); |
| } |
| .submit-panel h2 { |
| margin: 0; |
| font-size: clamp(30px, 3.5vw, 46px); |
| font-weight: 400; |
| letter-spacing: -.04em; |
| line-height: 1; |
| } |
| .submit-panel p { max-width: 720px; margin: 14px 0 0; color: var(--muted); font-size: 13px; } |
| .submit-panel .button { |
| padding: 12px 16px; |
| border-radius: 7px; |
| background: var(--signal); |
| color: #222; |
| text-decoration: none; |
| font-size: 13px; |
| white-space: nowrap; |
| } |
| .site-footer { |
| padding: 16px 0 28px; |
| display: flex; |
| justify-content: space-between; |
| gap: 24px; |
| border-top: 1px solid var(--line); |
| color: var(--muted); |
| font-size: 12px; |
| } |
| .site-footer a { text-decoration-color: var(--signal); text-decoration-thickness: 3px; } |
| @keyframes enter { |
| from { opacity: 0; transform: translateY(10px); } |
| to { opacity: 1; transform: none; } |
| } |
| @media (max-width: 820px) { |
| .shell { width: min(100% - 20px, 1280px); } |
| .submit-panel { margin-top: 48px; padding: 22px 18px; grid-template-columns: 1fr; } |
| .submit-panel .button { justify-self: start; } |
| .site-footer { flex-direction: column; } |
| } |
| @media (prefers-reduced-motion: reduce) { |
| html { scroll-behavior: auto; } |
| *, *::before, *::after { animation-duration: .01ms !important; animation-iteration-count: 1 !important; } |
| } |
| </style> |
| </head> |
| <body class="leaderboard-space"> |
| <main class="shell"> |
| <section class="leaderboard-section" id="results"> |
| <div class="tablebox"> |
| <table data-sortable-table> |
| <thead> |
| <tr> |
| <th aria-sort="none"> |
| <div class="column-actions"> |
| <button class="sort-button" type="button" data-sort-key="configuration" |
| data-sort-type="text" data-default-direction="ascending"> |
| Configuration <span class="sort-arrow" aria-hidden="true">↕</span> |
| </button> |
| </div> |
| </th> |
| <th class="numeric" aria-sort="none"> |
| <div class="column-actions"> |
| <button class="sort-button" type="button" data-sort-key="full_run_cost_usd" |
| data-sort-type="number" data-default-direction="ascending"> |
| Full-run cost <span class="sort-arrow" aria-hidden="true">↕</span> |
| </button> |
| <button class="definition-button" type="button" aria-label="Explain full-run cost" |
| aria-expanded="false" data-metric-title="Full-run cost" |
| data-metric-definition="Cost for all 32 documents using the best available run evidence; per-row hints explain each basis.">i</button> |
| </div> |
| </th> |
| <th class="numeric" aria-sort="none"> |
| <div class="column-actions"> |
| <button class="sort-button" type="button" data-sort-key="exact_record_recall" |
| data-sort-type="number" data-default-direction="descending"> |
| Exact recall <span class="sort-arrow" aria-hidden="true">↕</span> |
| </button> |
| <button class="definition-button" type="button" aria-label="Explain exact recall" |
| aria-expanded="false" data-metric-title="Exact recall" |
| data-metric-definition="Exact matched gold records divided by all gold records. Every normalized field in a record must match.">i</button> |
| </div> |
| </th> |
| <th class="numeric" aria-sort="descending"> |
| <div class="column-actions"> |
| <button class="sort-button" type="button" data-sort-key="complete_documents" |
| data-sort-type="number" data-default-direction="descending"> |
| Complete docs <span class="sort-arrow" aria-hidden="true">↓</span> |
| </button> |
| <button class="definition-button" type="button" aria-label="Explain complete documents" |
| aria-expanded="false" data-metric-title="Complete documents" |
| data-metric-definition="Documents where predicted and gold record multisets are identical, including duplicates and no extra records.">i</button> |
| </div> |
| </th> |
| <th class="numeric" aria-sort="none"> |
| <div class="column-actions"> |
| <button class="sort-button" type="button" data-sort-key="structural_exact_recall" |
| data-sort-type="number" data-default-direction="descending"> |
| Structural <span class="sort-arrow" aria-hidden="true">↕</span> |
| </button> |
| <button class="definition-button" type="button" aria-label="Explain structural recall" |
| aria-expanded="false" data-metric-title="Structural recall" |
| data-metric-definition="Exact-record recall on multisection, cross-page, multi-hop, and other structurally difficult document families.">i</button> |
| </div> |
| </th> |
| <th class="numeric" aria-sort="none"> |
| <div class="column-actions"> |
| <button class="sort-button" type="button" data-sort-key="scale_control_exact_recall" |
| data-sort-type="number" data-default-direction="descending"> |
| Scale ctrl <span class="sort-arrow" aria-hidden="true">↕</span> |
| </button> |
| <button class="definition-button" type="button" aria-label="Explain scale-control recall" |
| aria-expanded="false" data-metric-title="Scale-control recall" |
| data-metric-definition="Exact-record recall on dense schedules and returns that stress scale, OCR preservation, and output completeness.">i</button> |
| </div> |
| </th> |
| <th class="numeric" aria-sort="none"> |
| <div class="column-actions"> |
| <button class="sort-button" type="button" data-sort-key="weighted_f1" |
| data-sort-type="number" data-default-direction="descending"> |
| Field F1 <span class="sort-arrow" aria-hidden="true">↕</span> |
| </button> |
| <button class="definition-button" type="button" aria-label="Explain field F1" |
| aria-expanded="false" data-metric-title="Field F1" |
| data-metric-definition="F1 over matched flattened field-value pairs. A secondary partial-credit diagnostic for near misses.">i</button> |
| </div> |
| </th> |
| </tr> |
| </thead> |
| <tbody><tr data-result-row data-original-rank='1' data-details-row='result-details-1'><td class='configuration' data-column='configuration' data-sort-value='deep extract v3 (semantic tune)'><div class='model-line'><strong>Deep Extract v3 (semantic tune)</strong></div><span>Reducto · targeted fields · 2026-07-25 · n=1</span><div class='configuration-actions'><span class='protocol'>Raw PDF · prompt eng</span><button class="details-button" type="button" aria-expanded='false' aria-controls='result-details-1' data-model-name='Deep Extract v3 (semantic tune)' aria-label='Show document details for Deep Extract v3 (semantic tune)'><span data-details-label>Details</span><span class='details-symbol' aria-hidden='true'>+</span></button></div></td><td class='numeric price' data-column='full_run_cost_usd' data-sort-value='10.08'><span class='cost-cell'><span class='cost-value'>$10.08</span><span class='beta-badge'>beta</span><button class="definition-button cost-definition-button" type="button" aria-label="Explain full-run cost for Deep Extract v3 (semantic tune)" aria-expanded="false" data-metric-title="Full-run cost · Deep Extract v3 (semantic tune)" data-metric-definition="Experimental leaderboard cost of $10.08; current Reducto Standard list cost is $46.62." data-popover-template='pricing-comparison-1'>i</button><template id='pricing-comparison-1'><div class='pricing-comparison-popover'><p class='pricing-comparison'>Current Reducto Standard rate: $0.015/credit (observed 2026-07-28).</p><table class='pricing-comparison-table'><thead><tr><th>Pricing</th><th class='numeric'>Credits</th><th class='numeric'>Full-run cost</th></tr></thead><tbody><tr><th scope='row'>Current Standard</th><td class='numeric'>3,108.063</td><td class='numeric'>$46.62</td></tr><tr class='experimental'><th scope='row'>Experimental</th><td class='numeric'>≈672</td><td class='numeric'>$10.08</td></tr></tbody></table></div></template></span></td><td class='numeric' data-column='exact_record_recall' data-sort-value='0.959795939051995'><span class='metric-cell'><span class='metric-value'>96.0%</span></span></td><td class='numeric metric-best' data-column='complete_documents' data-sort-value='16'><span class='metric-cell'><span class='metric-value'>16/32</span></span></td><td class='numeric' data-column='structural_exact_recall' data-sort-value='0.8585690515806988'><span class='metric-cell'><span class='metric-value'>85.9%</span></span></td><td class='numeric metric-best' data-column='scale_control_exact_recall' data-sort-value='1.0'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric' data-column='weighted_f1' data-sort-value='0.9879272556680325'><span class='metric-cell'><span class='metric-value'>98.8%</span></span></td></tr><tr class='detail-row' id='result-details-1' data-detail-for='1' hidden><td colspan='7'><div class='detail-panel'><div class='detail-panel-heading'><strong>Deep Extract v3 (semantic tune)</strong><span>32 documents · n=1</span></div><div class='document-scroll'><table class='document-table'><thead><tr><th>Document</th><th class='numeric'>Gold records</th><th class='numeric'>Predicted</th><th class='numeric'>Exact recall</th><th class='numeric'>Field F1</th><th>Complete</th></tr></thead><tbody><tr><td class='document-name'>driver_mvr_packet_001</td><td class='numeric'>260</td><td class='numeric'>260</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_002</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_003</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>driver_schedule_sparse_001</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_001</td><td class='numeric'>1159</td><td class='numeric'>1159</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_002</td><td class='numeric'>2143</td><td class='numeric'>2143</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_003</td><td class='numeric'>2195</td><td class='numeric'>2195</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_004</td><td class='numeric'>2239</td><td class='numeric'>2239</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_005</td><td class='numeric'>2379</td><td class='numeric'>2379</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_006</td><td class='numeric'>2434</td><td class='numeric'>2434</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_007</td><td class='numeric'>2445</td><td class='numeric'>2445</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_008</td><td class='numeric'>2571</td><td class='numeric'>2571</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_001</td><td class='numeric'>335</td><td class='numeric'>335</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_002</td><td class='numeric'>461</td><td class='numeric'>475</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>92.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.7%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_001</td><td class='numeric'>582</td><td class='numeric'>558</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.9%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_002</td><td class='numeric'>998</td><td class='numeric'>962</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.4%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_005</td><td class='numeric'>1157</td><td class='numeric'>1115</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.4%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_001</td><td class='numeric'>650</td><td class='numeric'>649</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_002</td><td class='numeric'>650</td><td class='numeric'>649</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_001</td><td class='numeric'>760</td><td class='numeric'>760</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_002</td><td class='numeric'>760</td><td class='numeric'>760</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>loss_run_external_001</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_002</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_003</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>92.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_040_001_crosspage</td><td class='numeric'>40</td><td class='numeric'>40</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_cgl_040_001</td><td class='numeric'>562</td><td class='numeric'>581</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>24.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>81.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_012_001_crosspage</td><td class='numeric'>12</td><td class='numeric'>12</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.7%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_025_001_crosspage</td><td class='numeric'>25</td><td class='numeric'>25</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_bop_012_001</td><td class='numeric'>344</td><td class='numeric'>341</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>53.2%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.7%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_wc_025_001</td><td class='numeric'>438</td><td class='numeric'>456</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>28.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>89.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_001</td><td class='numeric'>800</td><td class='numeric'>800</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_002</td><td class='numeric'>800</td><td class='numeric'>800</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr></tbody></table></div><section class='prompt-panel'><div class='prompt-panel-heading'><strong>Prompt template</strong><span>Prompt engineered after observing benchmark failure modes.</span></div><section class='prompt-template'><strong>Generated field contract</strong><div class='prompt-scroll'><pre><code>Output JSON shape: |
| { "records": [ ... ] } |
|
|
| General requirements: |
| - Extract every target record visible in the document. |
| - Use the exact field names below; do not invent fields that are not listed. |
| {{RECORD_TYPE_REQUIREMENT}} |
| - If a listed field is not visible for a record, use an empty string. |
| - Preserve identifiers, dates, codes, locations, limits, rates, totals, names, and premiums exactly as shown. |
| - Do not deduplicate repeated records unless they are exact duplicate rows for the same item. |
|
|
| Allowed record groups and fields: |
| {{ALLOWED_RECORD_GROUPS_AND_FIELDS}} |
|
|
| All allowed fields: |
| {{ALL_ALLOWED_FIELDS}} |
|
|
| Target-scope rules: |
| - Extract records that match the listed fields, not every nearby note, notice, subtotal, or support table. |
| - Keep each field to its own value; do not include neighboring labels or carried-over header text. |
| - For tables split across pages or sections, preserve the inherited row context needed to complete each record. |
| - For policy packets, extract scheduled locations, classifications, coverages, forms, endorsements, material clauses, rating rows, and premium rows that match the allowed fields. |
| - For operations documents, extract one record per target table row or section-row combination matching the allowed fields. |
| {{CONDITIONAL_SCOPE_RULES}}</code></pre></div></section><section class='prompt-template'><strong>Prompt engineering additions</strong><div class='prompt-scroll'><pre><code>Field granularity and fidelity: |
| - Output only the atomic value each field names. Exclude surrounding labels, caption words that merely restate the field, header/breadcrumb prefixes, page-wrapper text, and annotations. |
| - Put each distinct attribute in its own field. Do not concatenate a second attribute (a category, unit, fuel/coverage qualifier, etc.) into another field's value. |
| - When the document shows both a full path/breadcrumb and a specific leaf label for a field, output the leaf label. |
| - When both a short canonical label and a longer restated narrative refer to the same field, output the short canonical value. |
| - Only output a value that is actually present for that record; if a listed field is not present, use an empty string. Do not infer, carry over, or fabricate. |
|
|
| Identifier and label fields: |
| - For fields that name an identifier or label (e.g. a schedule, return, form, or section |
| name), output only that identifier itself. Do not append a separate attribute — a fuel |
| type, coverage type, product category, or other section qualifier — onto it. Such |
| qualifiers are distinct attributes; put each in its own field when the schema provides |
| one, and otherwise omit it from the identifier value.</code></pre></div></section></section></div></td></tr><tr data-result-row data-original-rank='2' data-details-row='result-details-2'><td class='configuration' data-column='configuration' data-sort-value='deep extract v3 (strict contract)'><div class='model-line'><strong>Deep Extract v3 (strict contract)</strong></div><span>Reducto · identical contract · 2026-07-24 · n=1</span><div class='configuration-actions'><span class='protocol'>Raw PDF · strict contract</span><button class="details-button" type="button" aria-expanded='false' aria-controls='result-details-2' data-model-name='Deep Extract v3 (strict contract)' aria-label='Show document details for Deep Extract v3 (strict contract)'><span data-details-label>Details</span><span class='details-symbol' aria-hidden='true'>+</span></button></div></td><td class='numeric price' data-column='full_run_cost_usd' data-sort-value='11.02'><span class='cost-cell'><span class='cost-value'>$11.02</span><span class='beta-badge'>beta</span><button class="definition-button cost-definition-button" type="button" aria-label="Explain full-run cost for Deep Extract v3 (strict contract)" aria-expanded="false" data-metric-title="Full-run cost · Deep Extract v3 (strict contract)" data-metric-definition="Experimental leaderboard cost of $11.02; current Reducto Standard list cost is $46.49." data-popover-template='pricing-comparison-2'>i</button><template id='pricing-comparison-2'><div class='pricing-comparison-popover'><p class='pricing-comparison'>Current Reducto Standard rate: $0.015/credit (observed 2026-07-28).</p><table class='pricing-comparison-table'><thead><tr><th>Pricing</th><th class='numeric'>Credits</th><th class='numeric'>Full-run cost</th></tr></thead><tbody><tr><th scope='row'>Current Standard</th><td class='numeric'>3,099.268</td><td class='numeric'>$46.49</td></tr><tr class='experimental'><th scope='row'>Experimental</th><td class='numeric'>≈735</td><td class='numeric'>$11.02</td></tr></tbody></table></div></template></span></td><td class='numeric' data-column='exact_record_recall' data-sort-value='0.8895570796310687'><span class='metric-cell'><span class='metric-value'>89.0%</span></span></td><td class='numeric metric-best' data-column='complete_documents' data-sort-value='16'><span class='metric-cell'><span class='metric-value'>16/32</span></span></td><td class='numeric' data-column='structural_exact_recall' data-sort-value='0.6114808652246256'><span class='metric-cell'><span class='metric-value'>61.1%</span></span></td><td class='numeric metric-best' data-column='scale_control_exact_recall' data-sort-value='1.0'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric' data-column='weighted_f1' data-sort-value='0.9799063368944826'><span class='metric-cell'><span class='metric-value'>98.0%</span></span></td></tr><tr class='detail-row' id='result-details-2' data-detail-for='2' hidden><td colspan='7'><div class='detail-panel'><div class='detail-panel-heading'><strong>Deep Extract v3 (strict contract)</strong><span>32 documents · n=1</span></div><div class='document-scroll'><table class='document-table'><thead><tr><th>Document</th><th class='numeric'>Gold records</th><th class='numeric'>Predicted</th><th class='numeric'>Exact recall</th><th class='numeric'>Field F1</th><th>Complete</th></tr></thead><tbody><tr><td class='document-name'>driver_mvr_packet_001</td><td class='numeric'>260</td><td class='numeric'>260</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_002</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_003</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>driver_schedule_sparse_001</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_001</td><td class='numeric'>1159</td><td class='numeric'>1159</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_002</td><td class='numeric'>2143</td><td class='numeric'>2143</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_003</td><td class='numeric'>2195</td><td class='numeric'>2195</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_004</td><td class='numeric'>2239</td><td class='numeric'>2239</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_005</td><td class='numeric'>2379</td><td class='numeric'>2379</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_006</td><td class='numeric'>2434</td><td class='numeric'>2434</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_007</td><td class='numeric'>2445</td><td class='numeric'>2445</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_008</td><td class='numeric'>2571</td><td class='numeric'>2571</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_001</td><td class='numeric'>335</td><td class='numeric'>335</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_002</td><td class='numeric'>461</td><td class='numeric'>461</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_001</td><td class='numeric'>582</td><td class='numeric'>558</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.9%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_002</td><td class='numeric'>998</td><td class='numeric'>962</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>90.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_005</td><td class='numeric'>1157</td><td class='numeric'>1115</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>90.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_001</td><td class='numeric'>650</td><td class='numeric'>649</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_002</td><td class='numeric'>650</td><td class='numeric'>665</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_001</td><td class='numeric'>760</td><td class='numeric'>760</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_002</td><td class='numeric'>760</td><td class='numeric'>760</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>loss_run_external_001</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>80.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_002</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_003</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_040_001_crosspage</td><td class='numeric'>40</td><td class='numeric'>40</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_cgl_040_001</td><td class='numeric'>562</td><td class='numeric'>570</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>27.2%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>86.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_012_001_crosspage</td><td class='numeric'>12</td><td class='numeric'>12</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>87.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_025_001_crosspage</td><td class='numeric'>25</td><td class='numeric'>25</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_bop_012_001</td><td class='numeric'>344</td><td class='numeric'>338</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>39.2%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>94.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_wc_025_001</td><td class='numeric'>438</td><td class='numeric'>476</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>30.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>94.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_001</td><td class='numeric'>800</td><td class='numeric'>800</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_002</td><td class='numeric'>800</td><td class='numeric'>800</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr></tbody></table></div><section class='prompt-panel'><div class='prompt-panel-heading'><strong>Prompt template</strong><span>Primary Reducto comparison condition.</span></div><section class='prompt-template'><strong>Generated field contract</strong><div class='prompt-scroll'><pre><code>Output JSON shape: |
| { "records": [ ... ] } |
|
|
| General requirements: |
| - Extract every target record visible in the document. |
| - Use the exact field names below; do not invent fields that are not listed. |
| {{RECORD_TYPE_REQUIREMENT}} |
| - If a listed field is not visible for a record, use an empty string. |
| - Preserve identifiers, dates, codes, locations, limits, rates, totals, names, and premiums exactly as shown. |
| - Do not deduplicate repeated records unless they are exact duplicate rows for the same item. |
|
|
| Allowed record groups and fields: |
| {{ALLOWED_RECORD_GROUPS_AND_FIELDS}} |
|
|
| All allowed fields: |
| {{ALL_ALLOWED_FIELDS}} |
|
|
| Target-scope rules: |
| - Extract records that match the listed fields, not every nearby note, notice, subtotal, or support table. |
| - Keep each field to its own value; do not include neighboring labels or carried-over header text. |
| - For tables split across pages or sections, preserve the inherited row context needed to complete each record. |
| - For policy packets, extract scheduled locations, classifications, coverages, forms, endorsements, material clauses, rating rows, and premium rows that match the allowed fields. |
| - For operations documents, extract one record per target table row or section-row combination matching the allowed fields. |
| {{CONDITIONAL_SCOPE_RULES}}</code></pre></div></section></section></div></td></tr><tr data-result-row data-original-rank='3' data-details-row='result-details-3'><td class='configuration' data-column='configuration' data-sort-value='claude fable 5'><div class='model-line'><strong>Claude Fable 5</strong></div><span>Claude Code v2.1.216 · xhigh · 2026-07-21 · n=1</span><div class='configuration-actions'><span class='protocol'>Agentic CLI</span><button class="details-button" type="button" aria-expanded='false' aria-controls='result-details-3' data-model-name='Claude Fable 5' aria-label='Show document details for Claude Fable 5'><span data-details-label>Details</span><span class='details-symbol' aria-hidden='true'>+</span></button></div></td><td class='numeric price' data-column='full_run_cost_usd' data-sort-value='102.8426705'><span class='cost-cell'><span class='cost-value'>$102.84</span><button class="definition-button cost-definition-button" type="button" aria-label="Explain full-run cost for Claude Fable 5" aria-expanded="false" data-metric-title="Full-run cost · Claude Fable 5" data-metric-definition="Sum of 32 Claude Code API-equivalent estimates; subscription billing may differ.">i</button></span></td><td class='numeric' data-column='exact_record_recall' data-sort-value='0.9508429338829014'><span class='metric-cell'><span class='metric-value'>95.1%</span></span></td><td class='numeric' data-column='complete_documents' data-sort-value='9'><span class='metric-cell'><span class='metric-value'>9/32</span></span></td><td class='numeric' data-column='structural_exact_recall' data-sort-value='0.83990967435227'><span class='metric-cell'><span class='metric-value'>84.0%</span></span></td><td class='numeric' data-column='scale_control_exact_recall' data-sort-value='0.9949020533396271'><span class='metric-cell'><span class='metric-value'>99.5%</span></span></td><td class='numeric' data-column='weighted_f1' data-sort-value='0.967555883223247'><span class='metric-cell'><span class='metric-value'>96.8%</span></span></td></tr><tr class='detail-row' id='result-details-3' data-detail-for='3' hidden><td colspan='7'><div class='detail-panel'><div class='detail-panel-heading'><strong>Claude Fable 5</strong><span>32 documents · n=1 · Claude Code v2.1.216</span></div><div class='document-scroll'><table class='document-table'><thead><tr><th>Document</th><th class='numeric'>Gold records</th><th class='numeric'>Predicted</th><th class='numeric'>Exact recall</th><th class='numeric'>Field F1</th><th>Complete</th></tr></thead><tbody><tr><td class='document-name'>driver_mvr_packet_001</td><td class='numeric'>260</td><td class='numeric'>260</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_002</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_003</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>driver_schedule_sparse_001</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_001</td><td class='numeric'>1159</td><td class='numeric'>1159</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_002</td><td class='numeric'>2143</td><td class='numeric'>2143</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_003</td><td class='numeric'>2195</td><td class='numeric'>2195</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_004</td><td class='numeric'>2239</td><td class='numeric'>2239</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_005</td><td class='numeric'>2379</td><td class='numeric'>2379</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_006</td><td class='numeric'>2434</td><td class='numeric'>2434</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_007</td><td class='numeric'>2445</td><td class='numeric'>2445</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_008</td><td class='numeric'>2571</td><td class='numeric'>2571</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_001</td><td class='numeric'>335</td><td class='numeric'>335</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_002</td><td class='numeric'>461</td><td class='numeric'>461</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_001</td><td class='numeric'>582</td><td class='numeric'>558</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.9%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_002</td><td class='numeric'>998</td><td class='numeric'>962</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_005</td><td class='numeric'>1157</td><td class='numeric'>1115</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_001</td><td class='numeric'>650</td><td class='numeric'>649</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_002</td><td class='numeric'>650</td><td class='numeric'>649</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_001</td><td class='numeric'>760</td><td class='numeric'>760</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_002</td><td class='numeric'>760</td><td class='numeric'>760</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_001</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>88.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_002</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_003</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>90.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_040_001_crosspage</td><td class='numeric'>40</td><td class='numeric'>40</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_cgl_040_001</td><td class='numeric'>562</td><td class='numeric'>72</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>12.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>30.7%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_012_001_crosspage</td><td class='numeric'>12</td><td class='numeric'>12</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>multihop_025_001_crosspage</td><td class='numeric'>25</td><td class='numeric'>25</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>multihop_bop_012_001</td><td class='numeric'>344</td><td class='numeric'>100</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>18.6%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>41.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_wc_025_001</td><td class='numeric'>438</td><td class='numeric'>60</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>13.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>30.8%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_001</td><td class='numeric'>800</td><td class='numeric'>800</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_002</td><td class='numeric'>800</td><td class='numeric'>800</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr></tbody></table></div><section class='prompt-panel'><div class='prompt-panel-heading'><strong>Prompt template</strong><span>Repository-denied agentic run over released OCR transcripts. Generic operations and policy samples received a generated field contract; claim samples received the published incident JSON Schema.</span></div><section class='prompt-template'><strong>Task prompt — exact</strong><div class='prompt-scroll'><pre><code>You are dispatched as a subagent for a specific extraction task; skip superpowers skills. |
|
|
| Your workspace contains: |
| - document_ocr.md: OCR transcript of one benchmark PDF. |
| - field_contract.md: the target output shape and allowed fields. |
|
|
| The official ground truth is not present. Do not use files outside this workspace. |
|
|
| Task: |
| Extract the complete target list from document_ocr.md according to field_contract.md. |
| Write valid JSON to {{OUTPUT_FILE}}. |
|
|
| You may inspect the document, write temporary scripts, and verify counts. Before finishing: |
| - Confirm the JSON parses. |
| - Spot-check beginning, middle, and end records against the OCR. |
| - Ensure field names match field_contract.md exactly. |
| - Ensure every target row or target record type requested by the contract is represented. |
|
|
| Do not include explanations in the JSON file.</code></pre></div></section><section class='prompt-template'><strong>Field contract template — separate per-document input</strong><div class='prompt-scroll'><pre><code>Output JSON shape: |
| { "records": [ ... ] } |
|
|
| General requirements: |
| - Extract every target record visible in the document. |
| - Use the exact field names below; do not invent fields that are not listed. |
| {{RECORD_TYPE_REQUIREMENT}} |
| - If a listed field is not visible for a record, use an empty string. |
| - Preserve identifiers, dates, codes, locations, limits, rates, totals, names, and premiums exactly as shown. |
| - Do not deduplicate repeated records unless they are exact duplicate rows for the same item. |
|
|
| Allowed record groups and fields: |
| {{ALLOWED_RECORD_GROUPS_AND_FIELDS}} |
|
|
| All allowed fields: |
| {{ALL_ALLOWED_FIELDS}} |
|
|
| Target-scope rules: |
| - Extract records that match the listed fields, not every nearby note, notice, subtotal, or support table. |
| - Keep each field to its own value; do not include neighboring labels or carried-over header text. |
| - For tables split across pages or sections, preserve the inherited row context needed to complete each record. |
| - For policy packets, extract scheduled locations, classifications, coverages, forms, endorsements, material clauses, rating rows, and premium rows that match the allowed fields. |
| - For operations documents, extract one record per target table row or section-row combination matching the allowed fields. |
| {{CONDITIONAL_SCOPE_RULES}}</code></pre></div></section></section></div></td></tr><tr data-result-row data-original-rank='4' data-details-row='result-details-4'><td class='configuration' data-column='configuration' data-sort-value='gpt-5.6-sol'><div class='model-line'><strong>GPT-5.6-Sol</strong></div><span>Codex CLI v0.146.0 · xhigh · 2026-08-02 · n=3 · arithmetic mean</span><div class='configuration-actions'><span class='protocol'>Agentic CLI</span><button class="details-button" type="button" aria-expanded='false' aria-controls='result-details-4' data-model-name='GPT-5.6-Sol' aria-label='Show document details for GPT-5.6-Sol'><span data-details-label>Details</span><span class='details-symbol' aria-hidden='true'>+</span></button></div></td><td class='numeric price' data-column='full_run_cost_usd' data-sort-value='33.956977'><span class='cost-cell'><span class='cost-value'>$33.96</span><button class="definition-button cost-definition-button" type="button" aria-label="Explain full-run cost for GPT-5.6-Sol" aria-expanded="false" data-metric-title="Full-run cost · GPT-5.6-Sol" data-metric-definition="Mean API-equivalent cost across 3 saved full-corpus runs: $33.96 ± $1.20 sample SD.">i</button></span></td><td class='numeric metric-best' data-column='exact_record_recall' data-sort-value='0.9878036420149329'><span class='metric-cell'><span class='metric-value'>98.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for GPT-5.6-Sol" aria-expanded="false" data-metric-title="Exact recall variability · GPT-5.6-Sol" data-metric-definition="3-run arithmetic mean: 98.8% ± 0.2 pp sample SD.">i</button></span></td><td class='numeric' data-column='complete_documents' data-sort-value='8'><span class='metric-cell'><span class='metric-value'>8/32</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain complete documents variability for GPT-5.6-Sol" aria-expanded="false" data-metric-title="Complete documents variability · GPT-5.6-Sol" data-metric-definition="3-run arithmetic mean: 8/32 ± 1.0 documents sample SD.">i</button></span></td><td class='numeric metric-best' data-column='structural_exact_recall' data-sort-value='0.9700499168053245'><span class='metric-cell'><span class='metric-value'>97.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain structural recall variability for GPT-5.6-Sol" aria-expanded="false" data-metric-title="Structural recall variability · GPT-5.6-Sol" data-metric-definition="3-run arithmetic mean: 97.0% ± 0.6 pp sample SD.">i</button></span></td><td class='numeric' data-column='scale_control_exact_recall' data-sort-value='0.9948548501298088'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain scale-control recall variability for GPT-5.6-Sol" aria-expanded="false" data-metric-title="Scale-control recall variability · GPT-5.6-Sol" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric metric-best' data-column='weighted_f1' data-sort-value='0.997303631057584'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field f1 variability for GPT-5.6-Sol" aria-expanded="false" data-metric-title="Field F1 variability · GPT-5.6-Sol" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.0 pp sample SD.">i</button></span></td></tr><tr class='detail-row' id='result-details-4' data-detail-for='4' hidden><td colspan='7'><div class='detail-panel'><div class='detail-panel-heading'><strong>GPT-5.6-Sol</strong><span>32 documents · n=3 · Codex CLI v0.146.0</span></div><aside class='result-note'>The leaderboard averages three newer matched Sol runs made with Codex CLI v0.146.0, while the paper reports an older single-run baseline of 97.9%. Most of the difference comes from one large policy packet where the older run was unusually weak. Released transcripts and scoring targets match; a field-scope clarification affected three IFTA prompts but accounts for only a small part of the change. The run date and runtime also differ, so this is not a controlled model-improvement comparison.</aside><div class='document-scroll'><table class='document-table'><thead><tr><th>Document</th><th class='numeric'>Gold records</th><th class='numeric'>Predicted (mean)</th><th class='numeric'>Exact recall</th><th class='numeric'>Field F1</th><th>Complete runs</th></tr></thead><tbody><tr><td class='document-name'>driver_mvr_packet_001</td><td class='numeric'>260</td><td class='numeric'>262.7</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_001" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_001" data-metric-definition="3-run arithmetic mean: 99.0% ± 1.8 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_001" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_001" data-metric-definition="3-run arithmetic mean: 99.4% ± 1.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>2/3</span></td></tr><tr><td class='document-name'>driver_mvr_packet_002</td><td class='numeric'>500</td><td class='numeric'>505.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_002" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_002" data-metric-definition="3-run arithmetic mean: 98.9% ± 0.9 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_002" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_002" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>1/3</span></td></tr><tr><td class='document-name'>driver_mvr_packet_003</td><td class='numeric'>500</td><td class='numeric'>508.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_003" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_003" data-metric-definition="3-run arithmetic mean: 98.4% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_003" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_003" data-metric-definition="3-run arithmetic mean: 99.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>driver_schedule_sparse_001</td><td class='numeric'>500</td><td class='numeric'>500.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_schedule_sparse_001" aria-expanded="false" data-metric-title="Exact recall variability · driver_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_schedule_sparse_001" aria-expanded="false" data-metric-title="Field F1 variability · driver_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_001</td><td class='numeric'>1159</td><td class='numeric'>1159.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_001" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_002</td><td class='numeric'>2143</td><td class='numeric'>2143.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_002" data-metric-definition="3-run arithmetic mean: 96.5% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_002" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_003</td><td class='numeric'>2195</td><td class='numeric'>2195.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_003" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_003" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_003" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_003" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_004</td><td class='numeric'>2239</td><td class='numeric'>2239.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_004" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_004" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_004" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_004" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_005</td><td class='numeric'>2379</td><td class='numeric'>2379.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_005" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_005" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_005" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_005" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_006</td><td class='numeric'>2434</td><td class='numeric'>2434.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_006" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_006" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_006" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_006" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_007</td><td class='numeric'>2445</td><td class='numeric'>2445.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_007" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_007" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_007" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_007" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_008</td><td class='numeric'>2571</td><td class='numeric'>2571.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_008" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_008" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_008" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_008" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_multisection_return_001</td><td class='numeric'>335</td><td class='numeric'>335.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_multisection_return_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_multisection_return_001" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_multisection_return_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_multisection_return_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_multisection_return_002</td><td class='numeric'>461</td><td class='numeric'>461.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.1%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_multisection_return_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_multisection_return_002" data-metric-definition="3-run arithmetic mean: 97.1% ± 4.3 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_multisection_return_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_multisection_return_002" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.3 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_001</td><td class='numeric'>582</td><td class='numeric'>582.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_002</td><td class='numeric'>998</td><td class='numeric'>998.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_002" data-metric-definition="3-run arithmetic mean: 95.3% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_002" data-metric-definition="3-run arithmetic mean: 99.4% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_005</td><td class='numeric'>1157</td><td class='numeric'>1157.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_005" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_005" data-metric-definition="3-run arithmetic mean: 95.5% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_005" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_005" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_001</td><td class='numeric'>650</td><td class='numeric'>654.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_inquiry_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_inquiry_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_inquiry_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_inquiry_001" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_002</td><td class='numeric'>650</td><td class='numeric'>654.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_inquiry_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_inquiry_002" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_inquiry_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_inquiry_002" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_summary_001</td><td class='numeric'>760</td><td class='numeric'>760.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_summary_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_summary_001" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_summary_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_summary_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_summary_002</td><td class='numeric'>760</td><td class='numeric'>760.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_summary_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_summary_002" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_summary_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_summary_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_001</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>92.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_001" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_001" data-metric-definition="3-run arithmetic mean: 92.4% ± 5.4 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_001" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_001" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.4 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_002</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_002" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_002" data-metric-definition="3-run arithmetic mean: 97.8% ± 2.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_002" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_002" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_003</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_003" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_003" data-metric-definition="3-run arithmetic mean: 97.9% ± 0.2 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_003" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_003" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>mixed_040_001_crosspage</td><td class='numeric'>40</td><td class='numeric'>40.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for mixed_040_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · mixed_040_001_crosspage" data-metric-definition="3-run arithmetic mean: 97.5% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for mixed_040_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · mixed_040_001_crosspage" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>mixed_cgl_040_001</td><td class='numeric'>562</td><td class='numeric'>571.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for mixed_cgl_040_001" aria-expanded="false" data-metric-title="Exact recall variability · mixed_cgl_040_001" data-metric-definition="3-run arithmetic mean: 96.0% ± 6.9 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for mixed_cgl_040_001" aria-expanded="false" data-metric-title="Field F1 variability · mixed_cgl_040_001" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.3 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_012_001_crosspage</td><td class='numeric'>12</td><td class='numeric'>12.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_012_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · multihop_012_001_crosspage" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_012_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · multihop_012_001_crosspage" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>multihop_025_001_crosspage</td><td class='numeric'>25</td><td class='numeric'>25.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_025_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · multihop_025_001_crosspage" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_025_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · multihop_025_001_crosspage" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>multihop_bop_012_001</td><td class='numeric'>344</td><td class='numeric'>348.7</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>82.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_bop_012_001" aria-expanded="false" data-metric-title="Exact recall variability · multihop_bop_012_001" data-metric-definition="3-run arithmetic mean: 82.5% ± 0.2 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.1%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_bop_012_001" aria-expanded="false" data-metric-title="Field F1 variability · multihop_bop_012_001" data-metric-definition="3-run arithmetic mean: 98.1% ± 0.5 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_wc_025_001</td><td class='numeric'>438</td><td class='numeric'>470.7</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_wc_025_001" aria-expanded="false" data-metric-title="Exact recall variability · multihop_wc_025_001" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.4 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.2%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_wc_025_001" aria-expanded="false" data-metric-title="Field F1 variability · multihop_wc_025_001" data-metric-definition="3-run arithmetic mean: 98.2% ± 1.4 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_001</td><td class='numeric'>800</td><td class='numeric'>800.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for vehicle_schedule_sparse_001" aria-expanded="false" data-metric-title="Exact recall variability · vehicle_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for vehicle_schedule_sparse_001" aria-expanded="false" data-metric-title="Field F1 variability · vehicle_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_002</td><td class='numeric'>800</td><td class='numeric'>800.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for vehicle_schedule_sparse_002" aria-expanded="false" data-metric-title="Exact recall variability · vehicle_schedule_sparse_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for vehicle_schedule_sparse_002" aria-expanded="false" data-metric-title="Field F1 variability · vehicle_schedule_sparse_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr></tbody></table></div><section class='prompt-panel'><div class='prompt-panel-heading'><strong>Prompt template</strong><span>Repository-denied agentic run over released OCR transcripts. Generic operations and policy samples received a generated field contract; claim samples received the published incident JSON Schema.</span></div><section class='prompt-template'><strong>Task prompt — exact</strong><div class='prompt-scroll'><pre><code>You are dispatched as a subagent for a specific extraction task; skip superpowers skills. |
|
|
| Your workspace contains: |
| - document_ocr.md: OCR transcript of one benchmark PDF. |
| - field_contract.md: the target output shape and allowed fields. |
|
|
| The official ground truth is not present. Do not use files outside this workspace. |
|
|
| Task: |
| Extract the complete target list from document_ocr.md according to field_contract.md. |
| Write valid JSON to {{OUTPUT_FILE}}. |
|
|
| You may inspect the document, write temporary scripts, and verify counts. Before finishing: |
| - Confirm the JSON parses. |
| - Spot-check beginning, middle, and end records against the OCR. |
| - Ensure field names match field_contract.md exactly. |
| - Ensure every target row or target record type requested by the contract is represented. |
|
|
| Do not include explanations in the JSON file.</code></pre></div></section><section class='prompt-template'><strong>Field contract template — separate per-document input</strong><div class='prompt-scroll'><pre><code>Output JSON shape: |
| { "records": [ ... ] } |
|
|
| General requirements: |
| - Extract every target record visible in the document. |
| - Use the exact field names below; do not invent fields that are not listed. |
| {{RECORD_TYPE_REQUIREMENT}} |
| - If a listed field is not visible for a record, use an empty string. |
| - Preserve identifiers, dates, codes, locations, limits, rates, totals, names, and premiums exactly as shown. |
| - Do not deduplicate repeated records unless they are exact duplicate rows for the same item. |
|
|
| Allowed record groups and fields: |
| {{ALLOWED_RECORD_GROUPS_AND_FIELDS}} |
|
|
| All allowed fields: |
| {{ALL_ALLOWED_FIELDS}} |
|
|
| Target-scope rules: |
| - Extract records that match the listed fields, not every nearby note, notice, subtotal, or support table. |
| - Keep each field to its own value; do not include neighboring labels or carried-over header text. |
| - For tables split across pages or sections, preserve the inherited row context needed to complete each record. |
| - For policy packets, extract scheduled locations, classifications, coverages, forms, endorsements, material clauses, rating rows, and premium rows that match the allowed fields. |
| - For operations documents, extract one record per target table row or section-row combination matching the allowed fields. |
| {{CONDITIONAL_SCOPE_RULES}}</code></pre></div></section></section></div></td></tr><tr data-result-row data-original-rank='5' data-details-row='result-details-5'><td class='configuration' data-column='configuration' data-sort-value='claude opus 4.8'><div class='model-line'><strong>Claude Opus 4.8</strong></div><span>Claude Code v2.1.216 · xhigh · 2026-07-21 · n=1</span><div class='configuration-actions'><span class='protocol'>Agentic CLI</span><button class="details-button" type="button" aria-expanded='false' aria-controls='result-details-5' data-model-name='Claude Opus 4.8' aria-label='Show document details for Claude Opus 4.8'><span data-details-label>Details</span><span class='details-symbol' aria-hidden='true'>+</span></button></div></td><td class='numeric price' data-column='full_run_cost_usd' data-sort-value='44.86076025'><span class='cost-cell'><span class='cost-value'>$44.86</span><button class="definition-button cost-definition-button" type="button" aria-label="Explain full-run cost for Claude Opus 4.8" aria-expanded="false" data-metric-title="Full-run cost · Claude Opus 4.8" data-metric-definition="Sum of 32 Claude Code API-equivalent estimates; subscription billing may differ.">i</button></span></td><td class='numeric' data-column='exact_record_recall' data-sort-value='0.9770938207371871'><span class='metric-cell'><span class='metric-value'>97.7%</span></span></td><td class='numeric' data-column='complete_documents' data-sort-value='7'><span class='metric-cell'><span class='metric-value'>7/32</span></span></td><td class='numeric' data-column='structural_exact_recall' data-sort-value='0.9368908961255051'><span class='metric-cell'><span class='metric-value'>93.7%</span></span></td><td class='numeric' data-column='scale_control_exact_recall' data-sort-value='0.9930611281567147'><span class='metric-cell'><span class='metric-value'>99.3%</span></span></td><td class='numeric' data-column='weighted_f1' data-sort-value='0.9941570888224792'><span class='metric-cell'><span class='metric-value'>99.4%</span></span></td></tr><tr class='detail-row' id='result-details-5' data-detail-for='5' hidden><td colspan='7'><div class='detail-panel'><div class='detail-panel-heading'><strong>Claude Opus 4.8</strong><span>32 documents · n=1 · Claude Code v2.1.216</span></div><div class='document-scroll'><table class='document-table'><thead><tr><th>Document</th><th class='numeric'>Gold records</th><th class='numeric'>Predicted</th><th class='numeric'>Exact recall</th><th class='numeric'>Field F1</th><th>Complete</th></tr></thead><tbody><tr><td class='document-name'>driver_mvr_packet_001</td><td class='numeric'>260</td><td class='numeric'>268</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.9%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.1%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_002</td><td class='numeric'>500</td><td class='numeric'>508</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.4%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_003</td><td class='numeric'>500</td><td class='numeric'>508</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.4%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>driver_schedule_sparse_001</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_001</td><td class='numeric'>1159</td><td class='numeric'>1159</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_002</td><td class='numeric'>2143</td><td class='numeric'>2143</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>94.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.1%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_003</td><td class='numeric'>2195</td><td class='numeric'>2195</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_004</td><td class='numeric'>2239</td><td class='numeric'>2239</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_005</td><td class='numeric'>2379</td><td class='numeric'>2379</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_006</td><td class='numeric'>2434</td><td class='numeric'>2434</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_007</td><td class='numeric'>2445</td><td class='numeric'>2445</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_008</td><td class='numeric'>2571</td><td class='numeric'>2571</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_001</td><td class='numeric'>335</td><td class='numeric'>335</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_002</td><td class='numeric'>461</td><td class='numeric'>461</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_001</td><td class='numeric'>582</td><td class='numeric'>558</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.9%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_002</td><td class='numeric'>998</td><td class='numeric'>962</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_005</td><td class='numeric'>1157</td><td class='numeric'>1115</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_001</td><td class='numeric'>650</td><td class='numeric'>649</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_002</td><td class='numeric'>650</td><td class='numeric'>665</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_001</td><td class='numeric'>760</td><td class='numeric'>760</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_002</td><td class='numeric'>760</td><td class='numeric'>760</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_001</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_002</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_003</td><td class='numeric'>300</td><td class='numeric'>300</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.7%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_040_001_crosspage</td><td class='numeric'>40</td><td class='numeric'>40</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_cgl_040_001</td><td class='numeric'>562</td><td class='numeric'>562</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>multihop_012_001_crosspage</td><td class='numeric'>12</td><td class='numeric'>12</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>multihop_025_001_crosspage</td><td class='numeric'>25</td><td class='numeric'>25</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>multihop_bop_012_001</td><td class='numeric'>344</td><td class='numeric'>338</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>84.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.3%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_wc_025_001</td><td class='numeric'>438</td><td class='numeric'>438</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>31.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>92.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_001</td><td class='numeric'>800</td><td class='numeric'>800</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_002</td><td class='numeric'>800</td><td class='numeric'>800</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span></span></td><td class='document-complete'><span class='document-status complete'>Complete</span></td></tr></tbody></table></div><section class='prompt-panel'><div class='prompt-panel-heading'><strong>Prompt template</strong><span>Repository-denied agentic run over released OCR transcripts. Generic operations and policy samples received a generated field contract; claim samples received the published incident JSON Schema.</span></div><section class='prompt-template'><strong>Task prompt — exact</strong><div class='prompt-scroll'><pre><code>You are dispatched as a subagent for a specific extraction task; skip superpowers skills. |
|
|
| Your workspace contains: |
| - document_ocr.md: OCR transcript of one benchmark PDF. |
| - field_contract.md: the target output shape and allowed fields. |
|
|
| The official ground truth is not present. Do not use files outside this workspace. |
|
|
| Task: |
| Extract the complete target list from document_ocr.md according to field_contract.md. |
| Write valid JSON to {{OUTPUT_FILE}}. |
|
|
| You may inspect the document, write temporary scripts, and verify counts. Before finishing: |
| - Confirm the JSON parses. |
| - Spot-check beginning, middle, and end records against the OCR. |
| - Ensure field names match field_contract.md exactly. |
| - Ensure every target row or target record type requested by the contract is represented. |
|
|
| Do not include explanations in the JSON file.</code></pre></div></section><section class='prompt-template'><strong>Field contract template — separate per-document input</strong><div class='prompt-scroll'><pre><code>Output JSON shape: |
| { "records": [ ... ] } |
|
|
| General requirements: |
| - Extract every target record visible in the document. |
| - Use the exact field names below; do not invent fields that are not listed. |
| {{RECORD_TYPE_REQUIREMENT}} |
| - If a listed field is not visible for a record, use an empty string. |
| - Preserve identifiers, dates, codes, locations, limits, rates, totals, names, and premiums exactly as shown. |
| - Do not deduplicate repeated records unless they are exact duplicate rows for the same item. |
|
|
| Allowed record groups and fields: |
| {{ALLOWED_RECORD_GROUPS_AND_FIELDS}} |
|
|
| All allowed fields: |
| {{ALL_ALLOWED_FIELDS}} |
|
|
| Target-scope rules: |
| - Extract records that match the listed fields, not every nearby note, notice, subtotal, or support table. |
| - Keep each field to its own value; do not include neighboring labels or carried-over header text. |
| - For tables split across pages or sections, preserve the inherited row context needed to complete each record. |
| - For policy packets, extract scheduled locations, classifications, coverages, forms, endorsements, material clauses, rating rows, and premium rows that match the allowed fields. |
| - For operations documents, extract one record per target table row or section-row combination matching the allowed fields. |
| {{CONDITIONAL_SCOPE_RULES}}</code></pre></div></section></section></div></td></tr><tr data-result-row data-original-rank='6' data-details-row='result-details-6'><td class='configuration' data-column='configuration' data-sort-value='gpt-5.5'><div class='model-line'><strong>GPT-5.5</strong></div><span>Codex CLI v0.146.0 · xhigh · 2026-08-03 · n=3 · arithmetic mean</span><div class='configuration-actions'><span class='protocol'>Agentic CLI</span><button class="details-button" type="button" aria-expanded='false' aria-controls='result-details-6' data-model-name='GPT-5.5' aria-label='Show document details for GPT-5.5'><span data-details-label>Details</span><span class='details-symbol' aria-hidden='true'>+</span></button></div></td><td class='numeric price' data-column='full_run_cost_usd' data-sort-value='43.18508766666667'><span class='cost-cell'><span class='cost-value'>$43.19</span><button class="definition-button cost-definition-button" type="button" aria-label="Explain full-run cost for GPT-5.5" aria-expanded="false" data-metric-title="Full-run cost · GPT-5.5" data-metric-definition="Mean API-equivalent cost across 3 saved full-corpus runs: $43.19 ± $3.34 sample SD.">i</button></span></td><td class='numeric' data-column='exact_record_recall' data-sort-value='0.9656519927475028'><span class='metric-cell'><span class='metric-value'>96.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for GPT-5.5" aria-expanded="false" data-metric-title="Exact recall variability · GPT-5.5" data-metric-definition="3-run arithmetic mean: 96.6% ± 1.5 pp sample SD.">i</button></span></td><td class='numeric' data-column='complete_documents' data-sort-value='6.333333333333333'><span class='metric-cell'><span class='metric-value'>6/32</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain complete documents variability for GPT-5.5" aria-expanded="false" data-metric-title="Complete documents variability · GPT-5.5" data-metric-definition="3-run arithmetic mean: 6.3/32 ± 0.6 documents sample SD.">i</button></span></td><td class='numeric' data-column='structural_exact_recall' data-sort-value='0.8949766262578243'><span class='metric-cell'><span class='metric-value'>89.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain structural recall variability for GPT-5.5" aria-expanded="false" data-metric-title="Structural recall variability · GPT-5.5" data-metric-definition="3-run arithmetic mean: 89.5% ± 5.1 pp sample SD.">i</button></span></td><td class='numeric' data-column='scale_control_exact_recall' data-sort-value='0.9937219730941704'><span class='metric-cell'><span class='metric-value'>99.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain scale-control recall variability for GPT-5.5" aria-expanded="false" data-metric-title="Scale-control recall variability · GPT-5.5" data-metric-definition="3-run arithmetic mean: 99.4% ± 0.1 pp sample SD.">i</button></span></td><td class='numeric' data-column='weighted_f1' data-sort-value='0.992523637398692'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field f1 variability for GPT-5.5" aria-expanded="false" data-metric-title="Field F1 variability · GPT-5.5" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.3 pp sample SD.">i</button></span></td></tr><tr class='detail-row' id='result-details-6' data-detail-for='6' hidden><td colspan='7'><div class='detail-panel'><div class='detail-panel-heading'><strong>GPT-5.5</strong><span>32 documents · n=3 · Codex CLI v0.146.0</span></div><div class='document-scroll'><table class='document-table'><thead><tr><th>Document</th><th class='numeric'>Gold records</th><th class='numeric'>Predicted (mean)</th><th class='numeric'>Exact recall</th><th class='numeric'>Field F1</th><th>Complete runs</th></tr></thead><tbody><tr><td class='document-name'>driver_mvr_packet_001</td><td class='numeric'>260</td><td class='numeric'>176.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>67.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_001" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_001" data-metric-definition="3-run arithmetic mean: 67.7% ± 56.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>69.1%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_001" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_001" data-metric-definition="3-run arithmetic mean: 69.1% ± 53.5 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>2/3</span></td></tr><tr><td class='document-name'>driver_mvr_packet_002</td><td class='numeric'>500</td><td class='numeric'>500.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_002" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_002" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>driver_mvr_packet_003</td><td class='numeric'>500</td><td class='numeric'>500.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_003" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_003" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_003" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_003" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>driver_schedule_sparse_001</td><td class='numeric'>500</td><td class='numeric'>500.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_schedule_sparse_001" aria-expanded="false" data-metric-title="Exact recall variability · driver_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_schedule_sparse_001" aria-expanded="false" data-metric-title="Field F1 variability · driver_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_001</td><td class='numeric'>1159</td><td class='numeric'>1159.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_001" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_002</td><td class='numeric'>2143</td><td class='numeric'>2151.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_002" data-metric-definition="3-run arithmetic mean: 95.9% ± 1.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_002" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.3 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_003</td><td class='numeric'>2195</td><td class='numeric'>2195.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_003" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_003" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_003" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_003" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_004</td><td class='numeric'>2239</td><td class='numeric'>2227.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_004" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_004" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.9 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_004" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_004" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.5 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>2/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_005</td><td class='numeric'>2379</td><td class='numeric'>2379.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_005" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_005" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_005" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_005" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_006</td><td class='numeric'>2434</td><td class='numeric'>2434.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_006" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_006" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_006" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_006" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_007</td><td class='numeric'>2445</td><td class='numeric'>2445.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_007" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_007" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_007" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_007" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_008</td><td class='numeric'>2571</td><td class='numeric'>2571.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_008" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_008" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_008" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_008" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_multisection_return_001</td><td class='numeric'>335</td><td class='numeric'>335.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_multisection_return_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_multisection_return_001" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_multisection_return_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_multisection_return_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_multisection_return_002</td><td class='numeric'>461</td><td class='numeric'>461.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_multisection_return_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_multisection_return_002" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_multisection_return_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_multisection_return_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_001</td><td class='numeric'>582</td><td class='numeric'>582.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_001" data-metric-definition="3-run arithmetic mean: 95.9% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_001" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_002</td><td class='numeric'>998</td><td class='numeric'>998.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>63.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_002" data-metric-definition="3-run arithmetic mean: 63.5% ± 55.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_002" data-metric-definition="3-run arithmetic mean: 96.7% ± 4.7 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_005</td><td class='numeric'>1157</td><td class='numeric'>1157.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_005" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_005" data-metric-definition="3-run arithmetic mean: 95.5% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_005" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_005" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_001</td><td class='numeric'>650</td><td class='numeric'>659.7</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_inquiry_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_inquiry_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.2%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_inquiry_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_inquiry_001" data-metric-definition="3-run arithmetic mean: 99.2% ± 0.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_002</td><td class='numeric'>650</td><td class='numeric'>659.7</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_inquiry_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_inquiry_002" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.2%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_inquiry_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_inquiry_002" data-metric-definition="3-run arithmetic mean: 99.2% ± 0.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_summary_001</td><td class='numeric'>760</td><td class='numeric'>760.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_summary_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_summary_001" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_summary_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_summary_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_summary_002</td><td class='numeric'>760</td><td class='numeric'>760.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_summary_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_summary_002" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.2 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_summary_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_summary_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_001</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>92.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_001" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_001" data-metric-definition="3-run arithmetic mean: 92.3% ± 6.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_001" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_001" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.3 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_002</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>89.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_002" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_002" data-metric-definition="3-run arithmetic mean: 89.8% ± 0.2 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_002" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_002" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_003</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_003" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_003" data-metric-definition="3-run arithmetic mean: 91.9% ± 5.3 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_003" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_003" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.2 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>mixed_040_001_crosspage</td><td class='numeric'>40</td><td class='numeric'>40.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for mixed_040_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · mixed_040_001_crosspage" data-metric-definition="3-run arithmetic mean: 0.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>94.2%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for mixed_040_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · mixed_040_001_crosspage" data-metric-definition="3-run arithmetic mean: 94.2% ± 2.5 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>mixed_cgl_040_001</td><td class='numeric'>562</td><td class='numeric'>573.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>75.1%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for mixed_cgl_040_001" aria-expanded="false" data-metric-title="Exact recall variability · mixed_cgl_040_001" data-metric-definition="3-run arithmetic mean: 75.1% ± 14.6 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.1%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for mixed_cgl_040_001" aria-expanded="false" data-metric-title="Field F1 variability · mixed_cgl_040_001" data-metric-definition="3-run arithmetic mean: 97.1% ± 1.2 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_012_001_crosspage</td><td class='numeric'>12</td><td class='numeric'>12.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_012_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · multihop_012_001_crosspage" data-metric-definition="3-run arithmetic mean: 0.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>93.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_012_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · multihop_012_001_crosspage" data-metric-definition="3-run arithmetic mean: 93.3% ± 1.4 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_025_001_crosspage</td><td class='numeric'>25</td><td class='numeric'>25.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_025_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · multihop_025_001_crosspage" data-metric-definition="3-run arithmetic mean: 0.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_025_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · multihop_025_001_crosspage" data-metric-definition="3-run arithmetic mean: 95.7% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_bop_012_001</td><td class='numeric'>344</td><td class='numeric'>357.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>83.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_bop_012_001" aria-expanded="false" data-metric-title="Exact recall variability · multihop_bop_012_001" data-metric-definition="3-run arithmetic mean: 83.7% ± 1.8 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_bop_012_001" aria-expanded="false" data-metric-title="Field F1 variability · multihop_bop_012_001" data-metric-definition="3-run arithmetic mean: 97.7% ± 0.3 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_wc_025_001</td><td class='numeric'>438</td><td class='numeric'>470.7</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_wc_025_001" aria-expanded="false" data-metric-title="Exact recall variability · multihop_wc_025_001" data-metric-definition="3-run arithmetic mean: 98.9% ± 0.7 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_wc_025_001" aria-expanded="false" data-metric-title="Field F1 variability · multihop_wc_025_001" data-metric-definition="3-run arithmetic mean: 97.9% ± 1.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_001</td><td class='numeric'>800</td><td class='numeric'>800.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for vehicle_schedule_sparse_001" aria-expanded="false" data-metric-title="Exact recall variability · vehicle_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for vehicle_schedule_sparse_001" aria-expanded="false" data-metric-title="Field F1 variability · vehicle_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_002</td><td class='numeric'>800</td><td class='numeric'>800.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for vehicle_schedule_sparse_002" aria-expanded="false" data-metric-title="Exact recall variability · vehicle_schedule_sparse_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for vehicle_schedule_sparse_002" aria-expanded="false" data-metric-title="Field F1 variability · vehicle_schedule_sparse_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr></tbody></table></div><section class='prompt-panel'><div class='prompt-panel-heading'><strong>Prompt template</strong><span>Repository-denied agentic run over released OCR transcripts. Generic operations and policy samples received a generated field contract; claim samples received the published incident JSON Schema.</span></div><section class='prompt-template'><strong>Task prompt — exact</strong><div class='prompt-scroll'><pre><code>You are dispatched as a subagent for a specific extraction task; skip superpowers skills. |
|
|
| Your workspace contains: |
| - document_ocr.md: OCR transcript of one benchmark PDF. |
| - field_contract.md: the target output shape and allowed fields. |
|
|
| The official ground truth is not present. Do not use files outside this workspace. |
|
|
| Task: |
| Extract the complete target list from document_ocr.md according to field_contract.md. |
| Write valid JSON to {{OUTPUT_FILE}}. |
|
|
| You may inspect the document, write temporary scripts, and verify counts. Before finishing: |
| - Confirm the JSON parses. |
| - Spot-check beginning, middle, and end records against the OCR. |
| - Ensure field names match field_contract.md exactly. |
| - Ensure every target row or target record type requested by the contract is represented. |
|
|
| Do not include explanations in the JSON file.</code></pre></div></section><section class='prompt-template'><strong>Field contract template — separate per-document input</strong><div class='prompt-scroll'><pre><code>Output JSON shape: |
| { "records": [ ... ] } |
|
|
| General requirements: |
| - Extract every target record visible in the document. |
| - Use the exact field names below; do not invent fields that are not listed. |
| {{RECORD_TYPE_REQUIREMENT}} |
| - If a listed field is not visible for a record, use an empty string. |
| - Preserve identifiers, dates, codes, locations, limits, rates, totals, names, and premiums exactly as shown. |
| - Do not deduplicate repeated records unless they are exact duplicate rows for the same item. |
|
|
| Allowed record groups and fields: |
| {{ALLOWED_RECORD_GROUPS_AND_FIELDS}} |
|
|
| All allowed fields: |
| {{ALL_ALLOWED_FIELDS}} |
|
|
| Target-scope rules: |
| - Extract records that match the listed fields, not every nearby note, notice, subtotal, or support table. |
| - Keep each field to its own value; do not include neighboring labels or carried-over header text. |
| - For tables split across pages or sections, preserve the inherited row context needed to complete each record. |
| - For policy packets, extract scheduled locations, classifications, coverages, forms, endorsements, material clauses, rating rows, and premium rows that match the allowed fields. |
| - For operations documents, extract one record per target table row or section-row combination matching the allowed fields. |
| {{CONDITIONAL_SCOPE_RULES}}</code></pre></div></section></section></div></td></tr><tr data-result-row data-original-rank='7' data-details-row='result-details-7'><td class='configuration' data-column='configuration' data-sort-value='gpt-5.6-terra'><div class='model-line'><strong>GPT-5.6-Terra</strong></div><span>Codex CLI v0.146.0 · xhigh · 2026-07-31 · n=3 · arithmetic mean</span><div class='configuration-actions'><span class='protocol'>Agentic CLI</span><button class="details-button" type="button" aria-expanded='false' aria-controls='result-details-7' data-model-name='GPT-5.6-Terra' aria-label='Show document details for GPT-5.6-Terra'><span data-details-label>Details</span><span class='details-symbol' aria-hidden='true'>+</span></button></div></td><td class='numeric price' data-column='full_run_cost_usd' data-sort-value='15.0115356'><span class='cost-cell'><span class='cost-value'>$15.01</span><button class="definition-button cost-definition-button" type="button" aria-label="Explain full-run cost for GPT-5.6-Terra" aria-expanded="false" data-metric-title="Full-run cost · GPT-5.6-Terra" data-metric-definition="Mean API-equivalent cost across 3 saved full-corpus runs: $15.01 ± $0.41 sample SD.">i</button></span></td><td class='numeric' data-column='exact_record_recall' data-sort-value='0.9569917902631846'><span class='metric-cell'><span class='metric-value'>95.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for GPT-5.6-Terra" aria-expanded="false" data-metric-title="Exact recall variability · GPT-5.6-Terra" data-metric-definition="3-run arithmetic mean: 95.7% ± 2.0 pp sample SD.">i</button></span></td><td class='numeric' data-column='complete_documents' data-sort-value='6'><span class='metric-cell'><span class='metric-value'>6/32</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain complete documents variability for GPT-5.6-Terra" aria-expanded="false" data-metric-title="Complete documents variability · GPT-5.6-Terra" data-metric-definition="3-run arithmetic mean: 6/32 ± 1.0 documents sample SD.">i</button></span></td><td class='numeric' data-column='structural_exact_recall' data-sort-value='0.8646699944536883'><span class='metric-cell'><span class='metric-value'>86.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain structural recall variability for GPT-5.6-Terra" aria-expanded="false" data-metric-title="Structural recall variability · GPT-5.6-Terra" data-metric-definition="3-run arithmetic mean: 86.5% ± 7.3 pp sample SD.">i</button></span></td><td class='numeric' data-column='scale_control_exact_recall' data-sort-value='0.9936590354810794'><span class='metric-cell'><span class='metric-value'>99.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain scale-control recall variability for GPT-5.6-Terra" aria-expanded="false" data-metric-title="Scale-control recall variability · GPT-5.6-Terra" data-metric-definition="3-run arithmetic mean: 99.4% ± 0.1 pp sample SD.">i</button></span></td><td class='numeric' data-column='weighted_f1' data-sort-value='0.9922641184531067'><span class='metric-cell'><span class='metric-value'>99.2%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field f1 variability for GPT-5.6-Terra" aria-expanded="false" data-metric-title="Field F1 variability · GPT-5.6-Terra" data-metric-definition="3-run arithmetic mean: 99.2% ± 0.3 pp sample SD.">i</button></span></td></tr><tr class='detail-row' id='result-details-7' data-detail-for='7' hidden><td colspan='7'><div class='detail-panel'><div class='detail-panel-heading'><strong>GPT-5.6-Terra</strong><span>32 documents · n=3 · Codex CLI v0.146.0</span></div><div class='document-scroll'><table class='document-table'><thead><tr><th>Document</th><th class='numeric'>Gold records</th><th class='numeric'>Predicted (mean)</th><th class='numeric'>Exact recall</th><th class='numeric'>Field F1</th><th>Complete runs</th></tr></thead><tbody><tr><td class='document-name'>driver_mvr_packet_001</td><td class='numeric'>260</td><td class='numeric'>265.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_001" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_001" data-metric-definition="3-run arithmetic mean: 97.9% ± 1.8 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_001" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_001" data-metric-definition="3-run arithmetic mean: 98.7% ± 1.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>1/3</span></td></tr><tr><td class='document-name'>driver_mvr_packet_002</td><td class='numeric'>500</td><td class='numeric'>505.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_002" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_002" data-metric-definition="3-run arithmetic mean: 98.9% ± 0.9 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_002" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_002" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>1/3</span></td></tr><tr><td class='document-name'>driver_mvr_packet_003</td><td class='numeric'>500</td><td class='numeric'>500.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_003" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_003" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_003" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_003" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>driver_schedule_sparse_001</td><td class='numeric'>500</td><td class='numeric'>500.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_schedule_sparse_001" aria-expanded="false" data-metric-title="Exact recall variability · driver_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_schedule_sparse_001" aria-expanded="false" data-metric-title="Field F1 variability · driver_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_001</td><td class='numeric'>1159</td><td class='numeric'>1159.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_001" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_002</td><td class='numeric'>2143</td><td class='numeric'>2143.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_002" data-metric-definition="3-run arithmetic mean: 95.3% ± 1.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_002" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.3 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_003</td><td class='numeric'>2195</td><td class='numeric'>2195.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_003" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_003" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_003" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_003" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_004</td><td class='numeric'>2239</td><td class='numeric'>2239.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_004" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_004" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_004" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_004" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_005</td><td class='numeric'>2379</td><td class='numeric'>2379.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_005" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_005" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_005" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_005" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_006</td><td class='numeric'>2434</td><td class='numeric'>2434.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_006" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_006" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_006" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_006" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_007</td><td class='numeric'>2445</td><td class='numeric'>2445.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_007" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_007" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_007" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_007" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_008</td><td class='numeric'>2571</td><td class='numeric'>2571.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_008" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_008" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_008" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_008" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_multisection_return_001</td><td class='numeric'>335</td><td class='numeric'>335.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_multisection_return_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_multisection_return_001" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_multisection_return_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_multisection_return_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_multisection_return_002</td><td class='numeric'>461</td><td class='numeric'>461.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.1%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_multisection_return_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_multisection_return_002" data-metric-definition="3-run arithmetic mean: 97.1% ± 4.3 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_multisection_return_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_multisection_return_002" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.3 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_001</td><td class='numeric'>582</td><td class='numeric'>582.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_001" data-metric-definition="3-run arithmetic mean: 95.9% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_001" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_002</td><td class='numeric'>998</td><td class='numeric'>998.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>63.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_002" data-metric-definition="3-run arithmetic mean: 63.5% ± 55.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_002" data-metric-definition="3-run arithmetic mean: 96.7% ± 4.7 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_005</td><td class='numeric'>1157</td><td class='numeric'>1157.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>63.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_005" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_005" data-metric-definition="3-run arithmetic mean: 63.7% ± 55.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_005" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_005" data-metric-definition="3-run arithmetic mean: 96.8% ± 4.7 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_001</td><td class='numeric'>650</td><td class='numeric'>654.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_inquiry_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_inquiry_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_inquiry_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_inquiry_001" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_002</td><td class='numeric'>650</td><td class='numeric'>654.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_inquiry_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_inquiry_002" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_inquiry_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_inquiry_002" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_summary_001</td><td class='numeric'>760</td><td class='numeric'>760.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_summary_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_summary_001" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_summary_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_summary_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_summary_002</td><td class='numeric'>760</td><td class='numeric'>760.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_summary_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_summary_002" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_summary_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_summary_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_001</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>92.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_001" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_001" data-metric-definition="3-run arithmetic mean: 92.7% ± 5.8 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_001" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_001" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.4 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_002</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>93.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_002" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_002" data-metric-definition="3-run arithmetic mean: 93.7% ± 0.9 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_002" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_002" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_003</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>93.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_003" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_003" data-metric-definition="3-run arithmetic mean: 93.8% ± 0.8 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_003" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_003" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>mixed_040_001_crosspage</td><td class='numeric'>40</td><td class='numeric'>40.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for mixed_040_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · mixed_040_001_crosspage" data-metric-definition="3-run arithmetic mean: 0.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>92.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for mixed_040_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · mixed_040_001_crosspage" data-metric-definition="3-run arithmetic mean: 92.4% ± 1.7 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>mixed_cgl_040_001</td><td class='numeric'>562</td><td class='numeric'>570.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>86.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for mixed_cgl_040_001" aria-expanded="false" data-metric-title="Exact recall variability · mixed_cgl_040_001" data-metric-definition="3-run arithmetic mean: 86.6% ± 22.5 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for mixed_cgl_040_001" aria-expanded="false" data-metric-title="Field F1 variability · mixed_cgl_040_001" data-metric-definition="3-run arithmetic mean: 98.5% ± 2.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>1/3</span></td></tr><tr><td class='document-name'>multihop_012_001_crosspage</td><td class='numeric'>12</td><td class='numeric'>12.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_012_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · multihop_012_001_crosspage" data-metric-definition="3-run arithmetic mean: 0.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>92.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_012_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · multihop_012_001_crosspage" data-metric-definition="3-run arithmetic mean: 92.5% ± 1.4 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_025_001_crosspage</td><td class='numeric'>25</td><td class='numeric'>25.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_025_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · multihop_025_001_crosspage" data-metric-definition="3-run arithmetic mean: 0.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>94.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_025_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · multihop_025_001_crosspage" data-metric-definition="3-run arithmetic mean: 94.3% ± 2.5 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_bop_012_001</td><td class='numeric'>344</td><td class='numeric'>348.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>81.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_bop_012_001" aria-expanded="false" data-metric-title="Exact recall variability · multihop_bop_012_001" data-metric-definition="3-run arithmetic mean: 81.4% ± 1.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_bop_012_001" aria-expanded="false" data-metric-title="Field F1 variability · multihop_bop_012_001" data-metric-definition="3-run arithmetic mean: 97.7% ± 0.3 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_wc_025_001</td><td class='numeric'>438</td><td class='numeric'>483.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>93.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_wc_025_001" aria-expanded="false" data-metric-title="Exact recall variability · multihop_wc_025_001" data-metric-definition="3-run arithmetic mean: 93.6% ± 6.2 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_wc_025_001" aria-expanded="false" data-metric-title="Field F1 variability · multihop_wc_025_001" data-metric-definition="3-run arithmetic mean: 96.5% ± 1.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_001</td><td class='numeric'>800</td><td class='numeric'>800.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for vehicle_schedule_sparse_001" aria-expanded="false" data-metric-title="Exact recall variability · vehicle_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for vehicle_schedule_sparse_001" aria-expanded="false" data-metric-title="Field F1 variability · vehicle_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_002</td><td class='numeric'>800</td><td class='numeric'>800.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for vehicle_schedule_sparse_002" aria-expanded="false" data-metric-title="Exact recall variability · vehicle_schedule_sparse_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for vehicle_schedule_sparse_002" aria-expanded="false" data-metric-title="Field F1 variability · vehicle_schedule_sparse_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr></tbody></table></div><section class='prompt-panel'><div class='prompt-panel-heading'><strong>Prompt template</strong><span>Repository-denied agentic run over released OCR transcripts. Generic operations and policy samples received a generated field contract; claim samples received the published incident JSON Schema.</span></div><section class='prompt-template'><strong>Task prompt — exact</strong><div class='prompt-scroll'><pre><code>You are dispatched as a subagent for a specific extraction task; skip superpowers skills. |
|
|
| Your workspace contains: |
| - document_ocr.md: OCR transcript of one benchmark PDF. |
| - field_contract.md: the target output shape and allowed fields. |
|
|
| The official ground truth is not present. Do not use files outside this workspace. |
|
|
| Task: |
| Extract the complete target list from document_ocr.md according to field_contract.md. |
| Write valid JSON to {{OUTPUT_FILE}}. |
|
|
| You may inspect the document, write temporary scripts, and verify counts. Before finishing: |
| - Confirm the JSON parses. |
| - Spot-check beginning, middle, and end records against the OCR. |
| - Ensure field names match field_contract.md exactly. |
| - Ensure every target row or target record type requested by the contract is represented. |
|
|
| Do not include explanations in the JSON file.</code></pre></div></section><section class='prompt-template'><strong>Field contract template — separate per-document input</strong><div class='prompt-scroll'><pre><code>Output JSON shape: |
| { "records": [ ... ] } |
|
|
| General requirements: |
| - Extract every target record visible in the document. |
| - Use the exact field names below; do not invent fields that are not listed. |
| {{RECORD_TYPE_REQUIREMENT}} |
| - If a listed field is not visible for a record, use an empty string. |
| - Preserve identifiers, dates, codes, locations, limits, rates, totals, names, and premiums exactly as shown. |
| - Do not deduplicate repeated records unless they are exact duplicate rows for the same item. |
|
|
| Allowed record groups and fields: |
| {{ALLOWED_RECORD_GROUPS_AND_FIELDS}} |
|
|
| All allowed fields: |
| {{ALL_ALLOWED_FIELDS}} |
|
|
| Target-scope rules: |
| - Extract records that match the listed fields, not every nearby note, notice, subtotal, or support table. |
| - Keep each field to its own value; do not include neighboring labels or carried-over header text. |
| - For tables split across pages or sections, preserve the inherited row context needed to complete each record. |
| - For policy packets, extract scheduled locations, classifications, coverages, forms, endorsements, material clauses, rating rows, and premium rows that match the allowed fields. |
| - For operations documents, extract one record per target table row or section-row combination matching the allowed fields. |
| {{CONDITIONAL_SCOPE_RULES}}</code></pre></div></section></section></div></td></tr><tr data-result-row data-original-rank='8' data-details-row='result-details-8'><td class='configuration' data-column='configuration' data-sort-value='gpt-5.6-luna'><div class='model-line'><strong>GPT-5.6-Luna</strong></div><span>Codex CLI v0.146.0 · xhigh · 2026-08-01 · n=3 · arithmetic mean</span><div class='configuration-actions'><span class='protocol'>Agentic CLI</span><button class="details-button" type="button" aria-expanded='false' aria-controls='result-details-8' data-model-name='GPT-5.6-Luna' aria-label='Show document details for GPT-5.6-Luna'><span data-details-label>Details</span><span class='details-symbol' aria-hidden='true'>+</span></button></div></td><td class='numeric price' data-column='full_run_cost_usd' data-sort-value='2.4144322933333333'><span class='cost-cell'><span class='cost-value'>$2.41</span><button class="definition-button cost-definition-button" type="button" aria-label="Explain full-run cost for GPT-5.6-Luna" aria-expanded="false" data-metric-title="Full-run cost · GPT-5.6-Luna" data-metric-definition="Mean API-equivalent cost across 3 saved full-corpus runs: $2.41 ± $0.18 sample SD.">i</button></span></td><td class='numeric' data-column='exact_record_recall' data-sort-value='0.9440634255661791'><span class='metric-cell'><span class='metric-value'>94.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for GPT-5.6-Luna" aria-expanded="false" data-metric-title="Exact recall variability · GPT-5.6-Luna" data-metric-definition="3-run arithmetic mean: 94.4% ± 3.4 pp sample SD.">i</button></span></td><td class='numeric' data-column='complete_documents' data-sort-value='5'><span class='metric-cell'><span class='metric-value'>5/32</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain complete documents variability for GPT-5.6-Luna" aria-expanded="false" data-metric-title="Complete documents variability · GPT-5.6-Luna" data-metric-definition="3-run arithmetic mean: 5/32 ± 1.0 documents sample SD.">i</button></span></td><td class='numeric' data-column='structural_exact_recall' data-sort-value='0.8251723318279058'><span class='metric-cell'><span class='metric-value'>82.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain structural recall variability for GPT-5.6-Luna" aria-expanded="false" data-metric-title="Structural recall variability · GPT-5.6-Luna" data-metric-definition="3-run arithmetic mean: 82.5% ± 11.4 pp sample SD.">i</button></span></td><td class='numeric' data-column='scale_control_exact_recall' data-sort-value='0.9912831405868933'><span class='metric-cell'><span class='metric-value'>99.1%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain scale-control recall variability for GPT-5.6-Luna" aria-expanded="false" data-metric-title="Scale-control recall variability · GPT-5.6-Luna" data-metric-definition="3-run arithmetic mean: 99.1% ± 0.3 pp sample SD.">i</button></span></td><td class='numeric' data-column='weighted_f1' data-sort-value='0.986640921096695'><span class='metric-cell'><span class='metric-value'>98.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field f1 variability for GPT-5.6-Luna" aria-expanded="false" data-metric-title="Field F1 variability · GPT-5.6-Luna" data-metric-definition="3-run arithmetic mean: 98.7% ± 1.1 pp sample SD.">i</button></span></td></tr><tr class='detail-row' id='result-details-8' data-detail-for='8' hidden><td colspan='7'><div class='detail-panel'><div class='detail-panel-heading'><strong>GPT-5.6-Luna</strong><span>32 documents · n=3 · Codex CLI v0.146.0</span></div><div class='document-scroll'><table class='document-table'><thead><tr><th>Document</th><th class='numeric'>Gold records</th><th class='numeric'>Predicted (mean)</th><th class='numeric'>Exact recall</th><th class='numeric'>Field F1</th><th>Complete runs</th></tr></thead><tbody><tr><td class='document-name'>driver_mvr_packet_001</td><td class='numeric'>260</td><td class='numeric'>178.7</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>66.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_001" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_001" data-metric-definition="3-run arithmetic mean: 66.7% ± 55.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>68.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_001" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_001" data-metric-definition="3-run arithmetic mean: 68.5% ± 53.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>1/3</span></td></tr><tr><td class='document-name'>driver_mvr_packet_002</td><td class='numeric'>500</td><td class='numeric'>502.7</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_002" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_002" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.9 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_002" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_002" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>2/3</span></td></tr><tr><td class='document-name'>driver_mvr_packet_003</td><td class='numeric'>500</td><td class='numeric'>338.7</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>66.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_mvr_packet_003" aria-expanded="false" data-metric-title="Exact recall variability · driver_mvr_packet_003" data-metric-definition="3-run arithmetic mean: 66.7% ± 56.4 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>67.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_mvr_packet_003" aria-expanded="false" data-metric-title="Field F1 variability · driver_mvr_packet_003" data-metric-definition="3-run arithmetic mean: 67.6% ± 55.2 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>1/3</span></td></tr><tr><td class='document-name'>driver_schedule_sparse_001</td><td class='numeric'>500</td><td class='numeric'>500.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for driver_schedule_sparse_001" aria-expanded="false" data-metric-title="Exact recall variability · driver_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for driver_schedule_sparse_001" aria-expanded="false" data-metric-title="Field F1 variability · driver_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_001</td><td class='numeric'>1159</td><td class='numeric'>1159.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_001" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_002</td><td class='numeric'>2143</td><td class='numeric'>2143.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>94.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_002" data-metric-definition="3-run arithmetic mean: 94.7% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.1%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_002" data-metric-definition="3-run arithmetic mean: 99.1% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_003</td><td class='numeric'>2195</td><td class='numeric'>2195.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_003" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_003" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_003" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_003" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_004</td><td class='numeric'>2239</td><td class='numeric'>2239.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_004" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_004" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_004" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_004" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_005</td><td class='numeric'>2379</td><td class='numeric'>2379.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_005" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_005" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_005" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_005" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_006</td><td class='numeric'>2434</td><td class='numeric'>2434.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_006" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_006" data-metric-definition="3-run arithmetic mean: 98.5% ± 2.7 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_006" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_006" data-metric-definition="3-run arithmetic mean: 99.5% ± 0.9 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status partial'>2/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_007</td><td class='numeric'>2445</td><td class='numeric'>2445.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_007" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_007" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_007" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_007" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_008</td><td class='numeric'>2571</td><td class='numeric'>2571.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_mileage_by_vehicle_008" aria-expanded="false" data-metric-title="Exact recall variability · ifta_mileage_by_vehicle_008" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_mileage_by_vehicle_008" aria-expanded="false" data-metric-title="Field F1 variability · ifta_mileage_by_vehicle_008" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_multisection_return_001</td><td class='numeric'>335</td><td class='numeric'>335.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_multisection_return_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_multisection_return_001" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_multisection_return_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_multisection_return_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_multisection_return_002</td><td class='numeric'>461</td><td class='numeric'>461.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>94.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_multisection_return_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_multisection_return_002" data-metric-definition="3-run arithmetic mean: 94.6% ± 4.3 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_multisection_return_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_multisection_return_002" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.3 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_001</td><td class='numeric'>582</td><td class='numeric'>582.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>95.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_001" data-metric-definition="3-run arithmetic mean: 95.9% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_001" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_002</td><td class='numeric'>998</td><td class='numeric'>998.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>63.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_002" data-metric-definition="3-run arithmetic mean: 63.5% ± 55.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_002" data-metric-definition="3-run arithmetic mean: 96.8% ± 4.7 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_return_schedule_005</td><td class='numeric'>1157</td><td class='numeric'>1157.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>63.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_return_schedule_005" aria-expanded="false" data-metric-title="Exact recall variability · ifta_return_schedule_005" data-metric-definition="3-run arithmetic mean: 63.7% ± 55.1 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_return_schedule_005" aria-expanded="false" data-metric-title="Field F1 variability · ifta_return_schedule_005" data-metric-definition="3-run arithmetic mean: 96.8% ± 4.7 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_001</td><td class='numeric'>650</td><td class='numeric'>649.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_inquiry_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_inquiry_001" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_inquiry_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_inquiry_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_002</td><td class='numeric'>650</td><td class='numeric'>649.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_inquiry_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_inquiry_002" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_inquiry_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_inquiry_002" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_summary_001</td><td class='numeric'>760</td><td class='numeric'>760.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_summary_001" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_summary_001" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_summary_001" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_summary_001" data-metric-definition="3-run arithmetic mean: 99.9% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>ifta_tax_summary_002</td><td class='numeric'>760</td><td class='numeric'>760.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for ifta_tax_summary_002" aria-expanded="false" data-metric-title="Exact recall variability · ifta_tax_summary_002" data-metric-definition="3-run arithmetic mean: 99.6% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for ifta_tax_summary_002" aria-expanded="false" data-metric-title="Field F1 variability · ifta_tax_summary_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_001</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>88.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_001" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_001" data-metric-definition="3-run arithmetic mean: 88.8% ± 8.4 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.3%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_001" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_001" data-metric-definition="3-run arithmetic mean: 99.3% ± 0.5 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_002</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.1%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_002" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_002" data-metric-definition="3-run arithmetic mean: 96.1% ± 0.8 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_002" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_002" data-metric-definition="3-run arithmetic mean: 99.7% ± 0.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>loss_run_external_003</td><td class='numeric'>300</td><td class='numeric'>300.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for loss_run_external_003" aria-expanded="false" data-metric-title="Exact recall variability · loss_run_external_003" data-metric-definition="3-run arithmetic mean: 97.8% ± 0.4 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>99.8%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for loss_run_external_003" aria-expanded="false" data-metric-title="Field F1 variability · loss_run_external_003" data-metric-definition="3-run arithmetic mean: 99.8% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>mixed_040_001_crosspage</td><td class='numeric'>40</td><td class='numeric'>40.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for mixed_040_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · mixed_040_001_crosspage" data-metric-definition="3-run arithmetic mean: 0.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for mixed_040_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · mixed_040_001_crosspage" data-metric-definition="3-run arithmetic mean: 91.4% ± 0.1 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>mixed_cgl_040_001</td><td class='numeric'>562</td><td class='numeric'>581.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>81.2%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for mixed_cgl_040_001" aria-expanded="false" data-metric-title="Exact recall variability · mixed_cgl_040_001" data-metric-definition="3-run arithmetic mean: 81.2% ± 28.9 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for mixed_cgl_040_001" aria-expanded="false" data-metric-title="Field F1 variability · mixed_cgl_040_001" data-metric-definition="3-run arithmetic mean: 97.6% ± 2.6 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_012_001_crosspage</td><td class='numeric'>12</td><td class='numeric'>12.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_012_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · multihop_012_001_crosspage" data-metric-definition="3-run arithmetic mean: 0.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.6%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_012_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · multihop_012_001_crosspage" data-metric-definition="3-run arithmetic mean: 91.6% ± 0.2 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_025_001_crosspage</td><td class='numeric'>25</td><td class='numeric'>25.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_025_001_crosspage" aria-expanded="false" data-metric-title="Exact recall variability · multihop_025_001_crosspage" data-metric-definition="3-run arithmetic mean: 0.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>91.4%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_025_001_crosspage" aria-expanded="false" data-metric-title="Field F1 variability · multihop_025_001_crosspage" data-metric-definition="3-run arithmetic mean: 91.4% ± 0.2 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_bop_012_001</td><td class='numeric'>344</td><td class='numeric'>345.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>72.5%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_bop_012_001" aria-expanded="false" data-metric-title="Exact recall variability · multihop_bop_012_001" data-metric-definition="3-run arithmetic mean: 72.5% ± 13.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>97.2%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_bop_012_001" aria-expanded="false" data-metric-title="Field F1 variability · multihop_bop_012_001" data-metric-definition="3-run arithmetic mean: 97.2% ± 1.2 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>multihop_wc_025_001</td><td class='numeric'>438</td><td class='numeric'>483.3</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>88.7%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for multihop_wc_025_001" aria-expanded="false" data-metric-title="Exact recall variability · multihop_wc_025_001" data-metric-definition="3-run arithmetic mean: 88.7% ± 10.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.9%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for multihop_wc_025_001" aria-expanded="false" data-metric-title="Field F1 variability · multihop_wc_025_001" data-metric-definition="3-run arithmetic mean: 96.9% ± 1.7 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status incomplete'>0/3</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_001</td><td class='numeric'>800</td><td class='numeric'>800.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for vehicle_schedule_sparse_001" aria-expanded="false" data-metric-title="Exact recall variability · vehicle_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for vehicle_schedule_sparse_001" aria-expanded="false" data-metric-title="Field F1 variability · vehicle_schedule_sparse_001" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_002</td><td class='numeric'>800</td><td class='numeric'>800.0</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain exact recall variability for vehicle_schedule_sparse_002" aria-expanded="false" data-metric-title="Exact recall variability · vehicle_schedule_sparse_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>100.0%</span><button class="definition-button variability-definition-button" type="button" aria-label="Explain field F1 variability for vehicle_schedule_sparse_002" aria-expanded="false" data-metric-title="Field F1 variability · vehicle_schedule_sparse_002" data-metric-definition="3-run arithmetic mean: 100.0% ± 0.0 pp sample SD.">i</button></span></td><td class='document-complete'><span class='document-status complete'>3/3</span></td></tr></tbody></table></div><section class='prompt-panel'><div class='prompt-panel-heading'><strong>Prompt template</strong><span>Repository-denied agentic run over released OCR transcripts. Generic operations and policy samples received a generated field contract; claim samples received the published incident JSON Schema.</span></div><section class='prompt-template'><strong>Task prompt — exact</strong><div class='prompt-scroll'><pre><code>You are dispatched as a subagent for a specific extraction task; skip superpowers skills. |
|
|
| Your workspace contains: |
| - document_ocr.md: OCR transcript of one benchmark PDF. |
| - field_contract.md: the target output shape and allowed fields. |
|
|
| The official ground truth is not present. Do not use files outside this workspace. |
|
|
| Task: |
| Extract the complete target list from document_ocr.md according to field_contract.md. |
| Write valid JSON to {{OUTPUT_FILE}}. |
|
|
| You may inspect the document, write temporary scripts, and verify counts. Before finishing: |
| - Confirm the JSON parses. |
| - Spot-check beginning, middle, and end records against the OCR. |
| - Ensure field names match field_contract.md exactly. |
| - Ensure every target row or target record type requested by the contract is represented. |
|
|
| Do not include explanations in the JSON file.</code></pre></div></section><section class='prompt-template'><strong>Field contract template — separate per-document input</strong><div class='prompt-scroll'><pre><code>Output JSON shape: |
| { "records": [ ... ] } |
|
|
| General requirements: |
| - Extract every target record visible in the document. |
| - Use the exact field names below; do not invent fields that are not listed. |
| {{RECORD_TYPE_REQUIREMENT}} |
| - If a listed field is not visible for a record, use an empty string. |
| - Preserve identifiers, dates, codes, locations, limits, rates, totals, names, and premiums exactly as shown. |
| - Do not deduplicate repeated records unless they are exact duplicate rows for the same item. |
|
|
| Allowed record groups and fields: |
| {{ALLOWED_RECORD_GROUPS_AND_FIELDS}} |
|
|
| All allowed fields: |
| {{ALL_ALLOWED_FIELDS}} |
|
|
| Target-scope rules: |
| - Extract records that match the listed fields, not every nearby note, notice, subtotal, or support table. |
| - Keep each field to its own value; do not include neighboring labels or carried-over header text. |
| - For tables split across pages or sections, preserve the inherited row context needed to complete each record. |
| - For policy packets, extract scheduled locations, classifications, coverages, forms, endorsements, material clauses, rating rows, and premium rows that match the allowed fields. |
| - For operations documents, extract one record per target table row or section-row combination matching the allowed fields. |
| {{CONDITIONAL_SCOPE_RULES}}</code></pre></div></section></section></div></td></tr><tr data-result-row data-original-rank='9' data-details-row='result-details-9'><td class='configuration' data-column='configuration' data-sort-value='bonsai 27b'><div class='model-line'><strong>Bonsai 27B</strong></div><span>Local + Together AI · page pipeline · 2026-07-26 · n=1</span><div class='configuration-actions'><span class='protocol'>Page pipeline</span><button class="details-button" type="button" aria-expanded='false' aria-controls='result-details-9' data-model-name='Bonsai 27B' aria-label='Show document details for Bonsai 27B'><span data-details-label>Details</span><span class='details-symbol' aria-hidden='true'>+</span></button></div></td><td class='numeric price metric-best' data-column='full_run_cost_usd' data-sort-value='0.0'><span class='cost-cell'><span class='cost-value'>$0.00</span><button class="definition-button cost-definition-button" type="button" aria-label="Explain full-run cost for Bonsai 27B" aria-expanded="false" data-metric-title="Full-run cost · Bonsai 27B" data-metric-definition="Together hosted this run on an endpoint advertised as free on 2026-07-28; no dollar charge was recorded.">i</button></span></td><td class='numeric' data-column='exact_record_recall' data-sort-value='0.1673705192742998'><span class='metric-cell'><span class='metric-value'>16.7%</span></span></td><td class='numeric' data-column='complete_documents' data-sort-value='0'><span class='metric-cell'><span class='metric-value'>0/32</span></span></td><td class='numeric' data-column='structural_exact_recall' data-sort-value='0.12384121701925363'><span class='metric-cell'><span class='metric-value'>12.4%</span></span></td><td class='numeric' data-column='scale_control_exact_recall' data-sort-value='0.18465895680906302'><span class='metric-cell'><span class='metric-value'>18.5%</span></span></td><td class='numeric' data-column='weighted_f1' data-sort-value='0.40665444893832153'><span class='metric-cell'><span class='metric-value'>40.7%</span></span></td></tr><tr class='detail-row' id='result-details-9' data-detail-for='9' hidden><td colspan='7'><div class='detail-panel'><div class='detail-panel-heading'><strong>Bonsai 27B</strong><span>32 documents · n=1</span></div><div class='document-scroll'><table class='document-table'><thead><tr><th>Document</th><th class='numeric'>Gold records</th><th class='numeric'>Predicted</th><th class='numeric'>Exact recall</th><th class='numeric'>Field F1</th><th>Complete</th></tr></thead><tbody><tr><td class='document-name'>driver_mvr_packet_001</td><td class='numeric'>260</td><td class='numeric'>230</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>71.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>86.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_002</td><td class='numeric'>500</td><td class='numeric'>501</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>92.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>98.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>driver_mvr_packet_003</td><td class='numeric'>500</td><td class='numeric'>444</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>78.2%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>89.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>driver_schedule_sparse_001</td><td class='numeric'>500</td><td class='numeric'>500</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>93.4%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>96.8%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_001</td><td class='numeric'>1159</td><td class='numeric'>162</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>5.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>19.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_002</td><td class='numeric'>2143</td><td class='numeric'>1693</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>17.9%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>53.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_003</td><td class='numeric'>2195</td><td class='numeric'>666</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>4.3%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>23.2%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_004</td><td class='numeric'>2239</td><td class='numeric'>1676</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>30.6%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>62.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_005</td><td class='numeric'>2379</td><td class='numeric'>900</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>17.4%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>39.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_006</td><td class='numeric'>2434</td><td class='numeric'>1901</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>17.1%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>57.2%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_007</td><td class='numeric'>2445</td><td class='numeric'>855</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>14.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>35.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_mileage_by_vehicle_008</td><td class='numeric'>2571</td><td class='numeric'>1771</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>20.5%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>54.2%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_001</td><td class='numeric'>335</td><td class='numeric'>125</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>23.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_multisection_return_002</td><td class='numeric'>461</td><td class='numeric'>270</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>32.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_001</td><td class='numeric'>582</td><td class='numeric'>450</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.9%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>52.7%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_002</td><td class='numeric'>998</td><td class='numeric'>114</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>8.8%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_return_schedule_005</td><td class='numeric'>1157</td><td class='numeric'>84</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>5.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_001</td><td class='numeric'>650</td><td class='numeric'>488</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>17.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_inquiry_002</td><td class='numeric'>650</td><td class='numeric'>630</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>24.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_001</td><td class='numeric'>760</td><td class='numeric'>42</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>6.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>ifta_tax_summary_002</td><td class='numeric'>760</td><td class='numeric'>364</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>29.2%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_001</td><td class='numeric'>300</td><td class='numeric'>283</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>54.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_002</td><td class='numeric'>300</td><td class='numeric'>263</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>52.2%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>loss_run_external_003</td><td class='numeric'>300</td><td class='numeric'>293</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>54.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_040_001_crosspage</td><td class='numeric'>40</td><td class='numeric'>315</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>6.1%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>mixed_cgl_040_001</td><td class='numeric'>562</td><td class='numeric'>109</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>6.4%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_012_001_crosspage</td><td class='numeric'>12</td><td class='numeric'>107</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>4.6%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_025_001_crosspage</td><td class='numeric'>25</td><td class='numeric'>101</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>10.5%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_bop_012_001</td><td class='numeric'>344</td><td class='numeric'>90</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>10.0%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>multihop_wc_025_001</td><td class='numeric'>438</td><td class='numeric'>208</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>0.0%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>18.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_001</td><td class='numeric'>800</td><td class='numeric'>591</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>42.6%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>74.7%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr><tr><td class='document-name'>vehicle_schedule_sparse_002</td><td class='numeric'>800</td><td class='numeric'>320</td><td class='numeric'><span class='metric-cell'><span class='metric-value'>20.8%</span></span></td><td class='numeric'><span class='metric-cell'><span class='metric-value'>43.9%</span></span></td><td class='document-complete'><span class='document-status incomplete'>Incomplete</span></td></tr></tbody></table></div><section class='prompt-panel'><div class='prompt-panel-heading'><strong>Prompt template</strong><span>Two-stage page pipeline over released OCR transcripts.</span></div><section class='prompt-template'><strong>Page extraction</strong><div class='prompt-scroll'><pre><code>Extract target record evidence from exactly one OCR page. |
|
|
| Return every target row visible on the page. Also return partial rows when this |
| page contains fields whose other fields may be on another page. Each output row |
| is a compact array in the exact column order below. Use null for a field that is |
| not visible. Never use an empty string, zero, an empty array, or an empty object |
| as a placeholder. Return an empty array for a group absent from the page. Never |
| guess. Ignore summaries, totals, instructions, and unrelated tables. |
|
|
| Output groups and exact row column order: |
| {{OUTPUT_GROUPS_AND_COLUMN_ORDER}} |
|
|
| Return only the schema-constrained JSON object, minified on one line. |
|
|
| OCR page: |
| {{DOCUMENT_OCR_PAGE}}</code></pre></div></section><section class='prompt-template'><strong>Record reduction</strong><div class='prompt-scroll'><pre><code>Consolidate evidence for exactly one logical {{RECORD_TYPE}} record. |
|
|
| Use only the candidate fragments below. They were extracted independently from |
| single source pages. Resolve repeated or conflicting values from those |
| fragments. Return one record with every required field. Use an empty string, |
| null, empty list, or zero only when the fragments do not provide a value. Do |
| not create a different logical record. |
|
|
| Required fields: |
| {{REQUIRED_FIELDS}} |
|
|
| Expected record JSON Schema: |
| {{RECORD_SCHEMA}} |
|
|
| Candidate fragments: |
| {{CANDIDATE_FRAGMENTS}}</code></pre></div></section></section></div></td></tr></tbody> |
| </table> |
| </div> |
| </section> |
|
|
| <section class="submit-panel" id="submit"> |
| <div> |
| <h2>Submit a verified result</h2> |
| <p>Run the reference evaluator, then open a repository issue or Space discussion with |
| the evaluation report, run metadata, and per-sample predictions. Scores are reproduced |
| before publication.</p> |
| </div> |
| <a class="button" href="https://github.com/kaydotai/longlistbench/issues">Submission details →</a> |
| </section> |
|
|
| <footer class="site-footer"> |
| <span>LongListBench v2.2.1 · Open benchmark</span> |
| <span>Full tier, family, and stressor slices in <a href="leaderboard_data.json">leaderboard_data.json</a></span> |
| </footer> |
| </main> |
| <div class="metric-popover" id="metric-definition-popover" role="dialog" |
| aria-modal="false" tabindex="-1" hidden> |
| <span class="metric-popover-label">Metric definition</span> |
| <strong></strong> |
| <div class="metric-popover-copy"></div> |
| </div> |
| <script> |
| (() => { |
| const table = document.querySelector("[data-sortable-table]"); |
| if (!table) return; |
| const tbody = table.querySelector("tbody"); |
| const buttons = Array.from(table.querySelectorAll(".sort-button")); |
| const definitionButtons = Array.from(table.querySelectorAll(".definition-button")); |
| const detailButtons = Array.from(table.querySelectorAll(".details-button")); |
| const popover = document.querySelector("#metric-definition-popover"); |
| const popoverTitle = popover.querySelector("strong"); |
| const popoverCopy = popover.querySelector(".metric-popover-copy"); |
| const collator = new Intl.Collator(undefined, { numeric: true, sensitivity: "base" }); |
| let activeDefinitionButton = null; |
| |
| function closeDefinition() { |
| definitionButtons.forEach((button) => button.setAttribute("aria-expanded", "false")); |
| popover.hidden = true; |
| activeDefinitionButton = null; |
| } |
| |
| function setDetailsButton(button, expanded) { |
| button.setAttribute("aria-expanded", String(expanded)); |
| button.setAttribute( |
| "aria-label", |
| (expanded ? "Hide" : "Show") + " document details for " + button.dataset.modelName |
| ); |
| button.querySelector("[data-details-label]").textContent = expanded ? "Hide" : "Details"; |
| button.querySelector(".details-symbol").textContent = expanded ? "−" : "+"; |
| } |
| |
| function closeDetails() { |
| detailButtons.forEach((button) => { |
| const panel = document.getElementById(button.getAttribute("aria-controls")); |
| panel.hidden = true; |
| setDetailsButton(button, false); |
| }); |
| } |
| |
| function toggleDetails(button) { |
| closeDefinition(); |
| const panel = document.getElementById(button.getAttribute("aria-controls")); |
| const opening = panel.hidden; |
| closeDetails(); |
| if (!opening) return; |
| panel.hidden = false; |
| setDetailsButton(button, true); |
| } |
| |
| function showDefinition(button) { |
| if (!button.closest(".detail-row")) closeDetails(); |
| if (activeDefinitionButton === button && !popover.hidden) { |
| closeDefinition(); |
| return; |
| } |
| closeDefinition(); |
| popoverTitle.textContent = button.dataset.metricTitle; |
| const template = button.dataset.popoverTemplate |
| ? document.getElementById(button.dataset.popoverTemplate) |
| : null; |
| popoverCopy.replaceChildren(); |
| if (template) { |
| popoverCopy.appendChild(template.content.cloneNode(true)); |
| } else { |
| popoverCopy.textContent = button.dataset.metricDefinition; |
| } |
| popover.setAttribute("aria-label", button.dataset.metricTitle + " definition"); |
| popover.hidden = false; |
| const anchor = button.getBoundingClientRect(); |
| const bounds = popover.getBoundingClientRect(); |
| const left = Math.min( |
| window.innerWidth - bounds.width - 12, |
| Math.max(12, anchor.right - bounds.width) |
| ); |
| const below = anchor.bottom + 10; |
| const top = below + bounds.height <= window.innerHeight - 12 |
| ? below |
| : Math.max(12, anchor.top - bounds.height - 10); |
| popover.style.left = left + "px"; |
| popover.style.top = top + "px"; |
| button.setAttribute("aria-expanded", "true"); |
| activeDefinitionButton = button; |
| popover.focus({ preventScroll: true }); |
| } |
| |
| function sortTable(button, direction) { |
| const key = button.dataset.sortKey; |
| const type = button.dataset.sortType; |
| const rows = Array.from(tbody.querySelectorAll("[data-result-row]")); |
| closeDetails(); |
| rows.sort((leftRow, rightRow) => { |
| const leftCell = leftRow.querySelector('[data-column="' + key + '"]'); |
| const rightCell = rightRow.querySelector('[data-column="' + key + '"]'); |
| const left = leftCell.dataset.sortValue; |
| const right = rightCell.dataset.sortValue; |
| const leftMissing = leftCell.dataset.sortMissing === "true"; |
| const rightMissing = rightCell.dataset.sortMissing === "true"; |
| if (leftMissing !== rightMissing) return leftMissing ? 1 : -1; |
| const comparison = type === "number" |
| ? Number(left) - Number(right) |
| : collator.compare(left, right); |
| if (comparison === 0) { |
| return Number(leftRow.dataset.originalRank) - Number(rightRow.dataset.originalRank); |
| } |
| return direction === "ascending" ? comparison : -comparison; |
| }); |
| rows.forEach((row) => { |
| tbody.appendChild(row); |
| const detailRow = document.getElementById(row.dataset.detailsRow); |
| if (detailRow) tbody.appendChild(detailRow); |
| }); |
| |
| buttons.forEach((candidate) => { |
| const heading = candidate.closest("th"); |
| const active = candidate === button; |
| heading.setAttribute("aria-sort", active ? direction : "none"); |
| candidate.querySelector(".sort-arrow").textContent = active |
| ? (direction === "ascending" ? "↑" : "↓") |
| : "↕"; |
| }); |
| } |
| |
| buttons.forEach((button) => { |
| button.addEventListener("click", () => { |
| const current = button.closest("th").getAttribute("aria-sort"); |
| const direction = current === "ascending" |
| ? "descending" |
| : current === "descending" |
| ? "ascending" |
| : button.dataset.defaultDirection; |
| sortTable(button, direction); |
| }); |
| }); |
| |
| definitionButtons.forEach((button) => { |
| button.addEventListener("click", (event) => { |
| event.stopPropagation(); |
| showDefinition(button); |
| }); |
| }); |
| |
| detailButtons.forEach((button) => { |
| button.addEventListener("click", () => toggleDetails(button)); |
| }); |
| |
| document.addEventListener("click", (event) => { |
| if (!popover.hidden && !popover.contains(event.target)) closeDefinition(); |
| }); |
| document.addEventListener("keydown", (event) => { |
| if (event.key !== "Escape") return; |
| if (!popover.hidden) { |
| const button = activeDefinitionButton; |
| closeDefinition(); |
| button.focus(); |
| return; |
| } |
| const expandedDetails = detailButtons.find( |
| (button) => button.getAttribute("aria-expanded") === "true" |
| ); |
| if (expandedDetails) { |
| closeDetails(); |
| expandedDetails.focus(); |
| } |
| }); |
| window.addEventListener("resize", closeDefinition); |
| window.addEventListener("scroll", closeDefinition, true); |
| })(); |
| </script> |
| </body> |
| </html> |
|
|
