Docling layout block audit

original.pdf docling.json layout_blocks.json layout_blocks.tsv layout_blocks.jsonl excluded_blocks.html excluded_blocks.tsv final_body_blocks.tsv visual_assets.tsv visual_assets.json

Summary

{
  "blocks": 170,
  "included_in_body": 12,
  "gray_background_blocks": 4,
  "framed_blocks": 11,
  "by_role_guess": {
    "metadata": 10,
    "page_header": 23,
    "title_candidate": 2,
    "affiliation": 3,
    "reference": 53,
    "unknown_text": 8,
    "abstract_candidate": 2,
    "back_matter_heading": 4,
    "body_candidate_excluded": 24,
    "page_footer": 18,
    "caption": 4,
    "body_heading": 5,
    "body": 12,
    "page_margin_footer": 1,
    "back_matter_text": 1
  },
  "by_docling_label": {
    "text": 61,
    "page_header": 23,
    "section_header": 12,
    "page_footer": 18,
    "caption": 3,
    "list_item": 53
  },
  "by_body_decision_reason": {
    "first_page_metadata": 9,
    "docling_page_header": 15,
    "non_body_heading": 4,
    "front_matter_author_line": 1,
    "inside_front_matter": 25,
    "inline_abstract": 1,
    "inside_abstract": 1,
    "outside_body_flow": 3,
    "document_ui": 2,
    "early_supporting_information_link": 1,
    "docling_page_footer": 10,
    "outside_body_flow_caption": 3,
    "visual_caption_heading": 1,
    "before_body_started": 1,
    "body_before_non_intro_heading": 1,
    "body_heading": 3,
    "page_margin_footer": 1,
    "body": 11,
    "metadata_line": 1,
    "back_matter_heading": 1,
    "after_back_matter_stop": 75
  },
  "visual_assets": {
    "count": 5,
    "indexable_count": 5,
    "suppressed_count": 0,
    "by_type": {
      "figure": 4,
      "table": 1
    },
    "by_caption_source": {
      "direct_caption_ref": 3,
      "missing_caption": 1,
      "nearby_text_caption": 1
    },
    "by_duplicate_reason": {
      "": 5
    },
    "missing_caption_count": 1
  }
}

Diff Summary

{
  "parsed_text_blocks": 170,
  "final_body_blocks": 12,
  "excluded_blocks": {
    "count": 158,
    "by_reason": {
      "after_back_matter_stop": 75,
      "inside_front_matter": 25,
      "docling_page_header": 15,
      "docling_page_footer": 10,
      "first_page_metadata": 9,
      "non_body_heading": 4,
      "body_heading": 3,
      "outside_body_flow": 3,
      "outside_body_flow_caption": 3,
      "document_ui": 2,
      "back_matter_heading": 1,
      "before_body_started": 1,
      "early_supporting_information_link": 1,
      "front_matter_author_line": 1,
      "inline_abstract": 1,
      "inside_abstract": 1,
      "metadata_line": 1,
      "page_margin_footer": 1,
      "visual_caption_heading": 1
    },
    "by_role_guess": {
      "reference": 53,
      "body_candidate_excluded": 24,
      "page_header": 23,
      "page_footer": 18,
      "metadata": 10,
      "unknown_text": 8,
      "body_heading": 5,
      "back_matter_heading": 4,
      "caption": 4,
      "affiliation": 3,
      "abstract_candidate": 2,
      "title_candidate": 2,
      "back_matter_text": 1,
      "page_margin_footer": 1
    },
    "by_risk_level": {
      "low": 125,
      "high": 23,
      "medium": 10
    },
    "high_risk_count": 23,
    "medium_risk_count": 10
  },
  "char_counts": {
    "parsed_text_chars": 37374,
    "final_body_chars": 2073,
    "excluded_chars": 35301
  }
}

Truncation

{
  "truncated": true,
  "message": "Body extraction stopped at page 6 block #/texts/90: Author Contributions. 75 following text blocks were excluded as after_back_matter_stop.",
  "stop_trigger": {
    "ref": "#/texts/90",
    "page": 6,
    "layout_order": 94,
    "role_guess": "back_matter_heading",
    "body_decision_reason": "back_matter_heading",
    "text_preview": "Author Contributions"
  },
  "first_truncated_block": {
    "ref": "#/texts/91",
    "page": 6,
    "layout_order": 95,
    "role_guess": "unknown_text",
    "body_decision_reason": "after_back_matter_stop",
    "text_preview": "¶ J.R. and J.K.P. had equal contributions."
  },
  "truncated_block_count": 75,
  "truncated_pages": [
    6,
    7,
    8
  ],
  "by_role_guess": {
    "reference": 52,
    "page_header": 8,
    "page_footer": 6,
    "unknown_text": 3,
    "affiliation": 2,
    "back_matter_heading": 2,
    "back_matter_text": 1,
    "body_candidate_excluded": 1
  }
}

Page Overlays

Green = final body chunks. Orange/purple asset boxes = actual figure/table assets used by ingestion. Cyan dashed text boxes = text blocks consumed by those assets as caption continuations. Red STOP = truncation trigger. Red boxes = blocks after truncation.

Page 1

Page 2

Page 3

Page 4

Page 5

Page 6

Page 7

Page 8

Visual Assets

这里对齐真实图表资产提取链路。caption_source=embedded_table_cell 表示表注来自 Docling table cell，不会出现在 text block 审计差集里；caption_continuation_used_by_asset 表示某个 text block 已被图表 caption 吸收，不应按普通 metadata 解读。

Blocks