{
  "title": "SWE Resume Controlled PDF Format Benchmark",
  "version": "1.0",
  "published": "2026-07-23",
  "last_updated": "2026-07-23",
  "scope": "Literal field recovery and section-order extraction in pdfplumber and pypdf using controlled one-page resume fixtures.",
  "not_measured": [
    "Ranking or acceptance in a proprietary applicant tracking system",
    "Recruiter preference or interview outcomes",
    "Optical character recognition",
    "Semantic quality of extracted resume content"
  ],
  "expected_field_count": 14,
  "expected_section_order": [
    "SUMMARY",
    "EXPERIENCE",
    "PROJECTS",
    "SKILLS",
    "EDUCATION"
  ],
  "parsers": {
    "pdfplumber": "0.11.9",
    "pypdf": "6.10.0"
  },
  "results": [
    {
      "id": "single-column",
      "label": "Single-column semantic",
      "layout": "One text column with conventional sections",
      "file": "/research/resume-format-benchmark/fixtures/single-column.pdf",
      "average_field_recall_percent": 100,
      "section_order_preserved_by_both": true,
      "verdict": "Strong extraction",
      "parsers": {
        "pdfplumber": {
          "matched_fields": 14,
          "expected_fields": 14,
          "field_recall_percent": 100,
          "section_order_preserved": true,
          "extracted_characters": 855,
          "missing_fields": []
        },
        "pypdf": {
          "matched_fields": 14,
          "expected_fields": 14,
          "field_recall_percent": 100,
          "section_order_preserved": true,
          "extracted_characters": 856,
          "missing_fields": []
        }
      }
    },
    {
      "id": "two-column",
      "label": "Two-column sidebar",
      "layout": "Skills and education in a left sidebar; experience in the main column",
      "file": "/research/resume-format-benchmark/fixtures/two-column.pdf",
      "average_field_recall_percent": 96,
      "section_order_preserved_by_both": false,
      "verdict": "Fields recover; review order",
      "parsers": {
        "pdfplumber": {
          "matched_fields": 13,
          "expected_fields": 14,
          "field_recall_percent": 93,
          "section_order_preserved": false,
          "extracted_characters": 855,
          "missing_fields": [
            "BASc Computer Engineering"
          ]
        },
        "pypdf": {
          "matched_fields": 14,
          "expected_fields": 14,
          "field_recall_percent": 100,
          "section_order_preserved": false,
          "extracted_characters": 856,
          "missing_fields": []
        }
      }
    },
    {
      "id": "table-layout",
      "label": "Table-based layout",
      "layout": "Section labels and content stored in table cells",
      "file": "/research/resume-format-benchmark/fixtures/table-layout.pdf",
      "average_field_recall_percent": 100,
      "section_order_preserved_by_both": true,
      "verdict": "Strong extraction",
      "parsers": {
        "pdfplumber": {
          "matched_fields": 14,
          "expected_fields": 14,
          "field_recall_percent": 100,
          "section_order_preserved": true,
          "extracted_characters": 855,
          "missing_fields": []
        },
        "pypdf": {
          "matched_fields": 14,
          "expected_fields": 14,
          "field_recall_percent": 100,
          "section_order_preserved": true,
          "extracted_characters": 856,
          "missing_fields": []
        }
      }
    },
    {
      "id": "image-only",
      "label": "Image-only scan",
      "layout": "A raster image embedded in a PDF with no text layer",
      "file": "/research/resume-format-benchmark/fixtures/image-only.pdf",
      "average_field_recall_percent": 0,
      "section_order_preserved_by_both": false,
      "verdict": "Poor extraction",
      "parsers": {
        "pdfplumber": {
          "matched_fields": 0,
          "expected_fields": 14,
          "field_recall_percent": 0,
          "section_order_preserved": false,
          "extracted_characters": 0,
          "missing_fields": [
            "Alex Morgan",
            "alex.morgan@example.com",
            "(555) 010-2026",
            "Software Engineer",
            "Northstar Systems",
            "Beacon Labs",
            "Python",
            "TypeScript",
            "AWS",
            "Kubernetes",
            "PostgreSQL",
            "GraphQL",
            "University of Waterloo",
            "BASc Computer Engineering"
          ]
        },
        "pypdf": {
          "matched_fields": 0,
          "expected_fields": 14,
          "field_recall_percent": 0,
          "section_order_preserved": false,
          "extracted_characters": 0,
          "missing_fields": [
            "Alex Morgan",
            "alex.morgan@example.com",
            "(555) 010-2026",
            "Software Engineer",
            "Northstar Systems",
            "Beacon Labs",
            "Python",
            "TypeScript",
            "AWS",
            "Kubernetes",
            "PostgreSQL",
            "GraphQL",
            "University of Waterloo",
            "BASc Computer Engineering"
          ]
        }
      }
    }
  ]
}
