RaV-IDP: A Reconstruction-as-Validation Framework for Faithful Intelligent Document Processing
Fuente:
arXiv
Salvato in:
| Autore principale: | |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2026
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910167004610560 |
|---|---|
| author | Jha, Pritesh |
| author_facet | Jha, Pritesh |
| contents | Intelligent document processing pipelines extract structured entities (tables, images, and text) from documents for use in downstream systems such as knowledge bases, retrieval-augmented generation, and analytics. A persistent limitation of existing pipelines is that extraction output is produced without any intrinsic mechanism to verify whether it faithfully represents the source. Model-internal confidence scores measure inference certainty, not correspondence to the document, and extraction errors pass silently into downstream consumers.
We present Reconstruction as Validation (RaV-IDP), a document processing pipeline that introduces reconstruction as a first-class architectural component. After each entity is extracted, a dedicated reconstructor renders the extracted representation back into a form comparable to the original document region, and a comparator scores fidelity between the reconstruction and the unmodified source crop. This fidelity score is a grounded, label-free quality signal. When fidelity falls below a per-entity-type threshold, a structured GPT-4.1 vision fallback is triggered and the validation loop repeats. We enforce a bootstrap constraint: the comparator always anchors against the original document region, never against the extraction, preventing the validation from becoming circular.
We further propose a per-stage evaluation framework pairing each pipeline component with an appropriate benchmark. The code pipeline is publicly available at https://github.com/pritesh-2711/RaV-IDP for experimentation and use. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_23644 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | RaV-IDP: A Reconstruction-as-Validation Framework for Faithful Intelligent Document Processing Jha, Pritesh Computer Vision and Pattern Recognition Artificial Intelligence Intelligent document processing pipelines extract structured entities (tables, images, and text) from documents for use in downstream systems such as knowledge bases, retrieval-augmented generation, and analytics. A persistent limitation of existing pipelines is that extraction output is produced without any intrinsic mechanism to verify whether it faithfully represents the source. Model-internal confidence scores measure inference certainty, not correspondence to the document, and extraction errors pass silently into downstream consumers. We present Reconstruction as Validation (RaV-IDP), a document processing pipeline that introduces reconstruction as a first-class architectural component. After each entity is extracted, a dedicated reconstructor renders the extracted representation back into a form comparable to the original document region, and a comparator scores fidelity between the reconstruction and the unmodified source crop. This fidelity score is a grounded, label-free quality signal. When fidelity falls below a per-entity-type threshold, a structured GPT-4.1 vision fallback is triggered and the validation loop repeats. We enforce a bootstrap constraint: the comparator always anchors against the original document region, never against the extraction, preventing the validation from becoming circular. We further propose a per-stage evaluation framework pairing each pipeline component with an appropriate benchmark. The code pipeline is publicly available at https://github.com/pritesh-2711/RaV-IDP for experimentation and use. |
| title | RaV-IDP: A Reconstruction-as-Validation Framework for Faithful Intelligent Document Processing |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2604.23644 |