这里是差集:Docling 全量解析 text block 减去最终会进入正文的 block。优先人工检查 risk=high / medium。
layout_review.html excluded_blocks.json excluded_blocks.tsv final_body_blocks.tsv visual_assets.tsv original.pdf
{
"parsed_text_blocks": 0,
"final_body_blocks": 0,
"excluded_blocks": {
"count": 0,
"by_reason": {},
"by_role_guess": {},
"by_risk_level": {},
"high_risk_count": 0,
"medium_risk_count": 0
},
"char_counts": {
"parsed_text_chars": 0,
"final_body_chars": 0,
"excluded_chars": 0
}
}
{
"truncated": false,
"message": "",
"truncated_block_count": 0
}
这些框显示被排除块,同时叠加真实图表资产框。青色虚线表示该 text block 被图表资产 caption 吸收;红色 STOP 是截断触发点,红色框是截断后被排除的块。
这是实际图表资产输出,不是审计层重新推断。
| # | type | label | page | caption source | suppressed | duplicate reason | rescue reason | group | confidence | bbox | caption |
|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | figure | Docling Figure 1 | 1 | missing_caption | 0.55 | [53.73, 630.89, 234.39, 92.21] | |||||
| 2 | figure | Docling Figure 2 | 1 | missing_caption | 0.55 | [317.24, 529.91, 227.44, 163.82] | |||||
| 3 | figure | Docling Figure 3 | 2 | missing_caption | 0.55 | [49.75, 81.97, 235.07, 185.22] | |||||
| 4 | figure | Docling Figure 4 | 2 | missing_caption | 0.55 | [307.43, 83.21, 241.68, 163.11] | |||||
| 5 | table | Docling Table 5 | 2 | missing_caption | 0.55 | [48.08, 645.52, 240.51, 80.59] |
| risk | page | order | label | role | reason | parser reason | production usage | trunc | body region | region | bg | bbox | raw text | cleaned text |
|---|