Call for Papers

Submission Paths

Topics of Interest

We welcome submissions on a broad range of topics related to document intelligence and understanding, including but not limited to:

• Structure-aware modeling for layout, reading order, hierarchy, and cross-element relations
• Multimodal document understanding across text, layout, images, figures, charts, forms, and tables
• Reasoning over numerical, temporal, logical, tabular, and cross-page evidence
• Grounding, attribution, provenance, and hallucination mitigation in document QA and generation
• Robustness to OCR, parsing, layout, missing-cell, and extraction noise
• Long-document, multi-page, and multi-document reasoning
• Document retrieval, extraction, summarization, and source-linked generation
• Knowledge integration through schemas, ontologies, and knowledge graphs
• Benchmarks and diagnostic evaluation for document reasoning, grounding, robustness, uncertainty, and interaction
• Human-in-the-loop systems for verification, exploration, assistive reading, and decision support
• Multilingual, low-resource, and underrepresented language/script document understanding
• Scientific, legal, financial, healthcare, government, and enterprise document processing
• Privacy, consent, dataset governance, compliance, auditability, and deployment constraints

Contribution Types

• Long papers: up to 8 pages of content, plus unlimited references and appendices
• Short papers: up to 4 pages of content, plus unlimited references and appendices
• Position or theory papers
• Benchmark, dataset, and survey papers
• System papers and demos
• Reproducibility studies
• Negative results and diagnostic analyses

Requirements

Submissions must follow the official ACL style templates and be submitted as PDF files in standard two-column ACL format. Direct submissions must be anonymized for double-blind review.

Submissions should be self-contained. Supplementary material should be included in the appendix, and reviewers are not required to assess separate supplementary files.

Review Process

DocInsights submissions will be reviewed through OpenReview. Direct submissions will receive a double-blind review with at least 3 reviews plus meta-review, with decisions based on technical quality, relevance, clarity, ethical considerations, and fit to the workshop.

ARR commitments will be evaluated using existing ARR reviews and meta-review, together with relevance and fit to DocInsights.

Multiple Submission Policy

Archival submissions to DocInsights must not be under review at another conference or workshop during the DocInsights review period and must not substantially overlap with previously published work or work under review elsewhere.

Non-archival submissions may be submitted to other venues in the future, provided they comply with the policies of those venues.

Policies And Links

• Direct Submission: OpenReview
• ARR Commitment: OpenReview
• ACL Rolling Review: ARR CFP
• ACL Code of Ethics: ACL policy
• ACL Anti-Harassment Policy: ACL policy

Important Dates