Khondo: A Multimodal Benchmark for Document Packet Splitting of Bangla Forms

• URL: http://arxiv.org/abs/2607.21780v1
• Date: Thu, 23 Jul 2026 19:52:58 GMT
• Title: Khondo: A Multimodal Benchmark for Document Packet Splitting of Bangla Forms
• Abstract summary: Khondo is the first benchmark for document packet splitting on Bangladeshi government forms.<n>Unlike prior English and OCR-text-based datasets, Khondo is bilingual (Bangla--English) and vision-native.<n>It spans five concatenation schemes, from sequential to fully shuffled, across 14 administrative domains.
• Score: 13.071620971838316
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Document packets, multiple documents concatenated into a single file, are common in government and administrative workflows, yet splitting them into their constituent documents is difficult, especially for low-resource languages. We introduce Khondo (Bangla for split/segment), the first benchmark for document packet splitting on Bangladeshi government forms. Unlike prior English and OCR-text-based datasets, Khondo is bilingual (Bangla--English) and vision-native; where models operate directly on page images. It spans five concatenation schemes, from sequential to fully shuffled, across 14 administrative domains, with ground-truth boundaries, domain types, and page order. Zero-shot evaluation of MLLMs shows they cluster pages into their source documents fairly well but struggle in restoring the original page order once shuffled. To isolate what drives this difficulty, we run two controlled analyses, varying the prompt instruction and then the packet language. Both primarily affect ordering rather than clustering: (a) explicit page-order instructions are necessary but insufficient, and (b) English packets are ordered more reliably than Bangla, making page arrangement the dominant challenge and language a secondary but consistent factor. Khondo establishes page-order reconstruction as a key open problem in vision-based, low-resource document understanding, and provides a controlled benchmark for measuring progress toward solving it. Our dataset and code is available at https://huggingface.co/datasets/Mausul/khondo

Related papers

• EMBLEM: Enhancing Multi-script Table Detection through Masking [14.822713626646332]
  We introduce MANDALA (Multi-script Annotated Documents for Table Detection), a dataset of 2,323 table-containing pages spanning 18 languages and 15 scripts across diverse domains.<n>We also propose EMBLEM, a masking-based paradigm for Multi-script Table Detection (MTD)<n>EMBLEM generates masked images that conceal script- and font-specific details, enabling models pre-trained on abundant English documents to focus on script-agnostic page layout.
  arXiv  Detail & Related papers  (2026-09-08T06:57:50Z)
• TongGuOCR: A Layout-Aware and Token-Augmented OCR MLLM for Chinese Historical Documents [50.64307290680222]
  TongGuOCR is a layout-aware and token-augmented multimodal large language model (MLLM) for OCR of Chinese historical documents.<n>TongGuOCR achieves 93.76 AR and reduces NED from 10.43 to 6.15 and RO-ED from 7.53 to 3.49 relative to the best competing score for each metric.
  arXiv  Detail & Related papers  (2026-08-08T04:50:00Z)
• BaFCo: A Document Understanding Benchmark for Complex Bangla Form Comprehension [12.574173764520728]
  BaFCo is a benchmark dataset for Bangla form comprehension.<n>BaFCo curates 200 multi-page complex Bangladeshi government forms.<n>We evaluate the latest MLLMs from the ChatGPT, Gemini, Claude, Qwen, and Kimi series.
  arXiv  Detail & Related papers  (2026-07-06T20:17:38Z)
• MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing [74.84107522458798]
  MPDocBench-Parse is a benchmark for multi-page document parsing in real-world applications.<n>It contains 433 manually annotated documents with 3,246 pages, covering 15 document types in English and Chinese, with diverse layout styles.
  arXiv  Detail & Related papers  (2026-05-21T07:36:41Z)
• MoDora: Tree-Based Semi-Structured Document Analysis System [62.01015188258797]
  Semi-structured documents integrate diverse interleaved data elements arranged in various and often irregular layouts.<n>MoDora is an LLM-powered system for semi-structured document analysis.<n> Experiments show MoDora outperforms baselines by 5.97%-61.07% in accuracy.
  arXiv  Detail & Related papers  (2026-02-26T14:48:49Z)
• KH-FUNSD: A Hierarchical and Fine-Grained Layout Analysis Dataset for Low-Resource Khmer Business Document [11.302542266122579]
  Khmer is a language spoken daily by over 17 million people in Cambodia.<n>Lack of dedicated resources is particularly acute for business documents.<n>We present textbfKH-FUNSD, the first publicly available dataset for Khmer form document understanding.
  arXiv  Detail & Related papers  (2025-12-04T13:28:44Z)
• Contextual Document Embeddings [77.22328616983417]
  We propose two complementary methods for contextualized document embeddings. First, an alternative contrastive learning objective that explicitly incorporates the document neighbors into the intra-batch contextual loss. Second, a new contextual architecture that explicitly encodes neighbor document information into the encoded representation.
  arXiv  Detail & Related papers  (2024-10-03T14:33:34Z)
• mPLUG-DocOwl2: High-resolution Compressing for OCR-free Multi-page Document Understanding [103.05835688963947]
  We propose a High-resolution DocCompressor module to compress each high-resolution document image into 324 tokens. DocOwl2 sets a new state-of-the-art across multi-page document understanding benchmarks and reduces first token latency by more than 50%. Compared to single-image MLLMs trained on similar data, our DocOwl2 achieves comparable single-page understanding performance with less than 20% of the visual tokens.
  arXiv  Detail & Related papers  (2024-09-05T11:09:00Z)
• DAPR: A Benchmark on Document-Aware Passage Retrieval [57.45793782107218]
  We propose and name this task emphDocument-Aware Passage Retrieval (DAPR) While analyzing the errors of the State-of-The-Art (SoTA) passage retrievers, we find the major errors (53.5%) are due to missing document context. Our created benchmark enables future research on developing and comparing retrieval systems for the new task.
  arXiv  Detail & Related papers  (2023-05-23T10:39:57Z)
• Cross-Modal Entity Matching for Visually Rich Documents [4.8119678510491815]
  Visually rich documents utilize visual cues to augment their semantics. Existing works that enable structured querying on these documents do not take this into account. We propose Juno -- a cross-modal entity matching framework to address this limitation.
  arXiv  Detail & Related papers  (2023-03-01T18:26:14Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.