{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T19:07:35Z","timestamp":1779131255688,"version":"3.51.4"},"reference-count":35,"publisher":"Association for Computing Machinery (ACM)","issue":"3","funder":[{"DOI":"10.13039\/501100003816","name":"National Key R\\\\&D Program of China","doi-asserted-by":"publisher","award":["2023YFB4502400"],"award-info":[{"award-number":["2023YFB4502400"]}],"id":[{"id":"10.13039\/501100003816","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Fundamental and Interdisciplinary Disciplines Breakthrough Plan of the Ministry of Education of China","award":["JYB2025XDXM103"],"award-info":[{"award-number":["JYB2025XDXM103"]}]},{"name":"Shenzhen Project","award":["CJGJZD20230724093403007"],"award-info":[{"award-number":["CJGJZD20230724093403007"]}]},{"DOI":"10.13039\/501100001809","name":"NSF of China","doi-asserted-by":"crossref","award":["62441236, 62372296, 62432007, U25A6024, 62502304, U25A20437, 62525202, 62232009"],"award-info":[{"award-number":["62441236, 62372296, 62432007, U25A6024, 62502304, U25A20437, 62525202, 62232009"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Proc. ACM Manag. Data"],"published-print":{"date-parts":[[2026,5,18]]},"abstract":"<jats:p>Semi-structured documents integrate diverse interleaved data elements (e.g., tables, charts, hierarchical paragraphs) arranged in various and often irregular layouts. These documents are widely observed across domains and account for a large portion of real-world data. However, existing methods struggle to support natural language question answering over these documents due to three main technical challenges: (1) The elements extracted by techniques like OCR are often fragmented and stripped of their original semantic context, making them inadequate for analysis. (2) Existing approaches lack effective representations to capture hierarchical structures within documents (e.g., associating tables with nested chapter titles) and to preserve layout-specific distinctions (e.g., differentiating sidebars from main content). (3) Answering questions often requires retrieving and aligning relevant information scattered across multiple regions or pages, such as linking a descriptive paragraph to table cells located elsewhere in the document.<\/jats:p>\n                  <jats:p>To address these issues, we propose Modora, an LLM-powered system for semi-structured document analysis. First, we adopt a local-alignment aggregation strategy to convert OCR-parsed elements into layout-aware components, and conduct type-specific information extraction for components with hierarchical titles or non-text elements. Second, we design the Component-Correlation Tree (CCTree) to hierarchically organize components, explicitly modeling inter-component relations and layout distinctions through a bottom-up cascade summarization process. Finally, we propose a question-type-aware retrieval strategy that supports (1) layout-based grid partitioning for location-based retrieval and (2) LLM-guided pruning for semantic-based retrieval. Experiments show Modora outperforms baselines by 5.97%-61.07% in accuracy. The code is at https:\/\/github.com\/weAIDB\/MoDora.<\/jats:p>","DOI":"10.1145\/3802089","type":"journal-article","created":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T18:19:16Z","timestamp":1779128356000},"page":"1-26","source":"Crossref","is-referenced-by-count":0,"title":["MoDora: Tree-Based Semi-Structured Document Analysis System"],"prefix":"10.1145","volume":"4","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-4489-9048","authenticated-orcid":false,"given":"Bangrui","family":"Xu","sequence":"first","affiliation":[{"name":"Shanghai Jiao Tong University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-0282-8629","authenticated-orcid":false,"given":"Qihang","family":"Yao","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-3449-4248","authenticated-orcid":false,"given":"Zirui","family":"Tang","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2285-7836","authenticated-orcid":false,"given":"Xuanhe","family":"Zhou","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2824-5299","authenticated-orcid":false,"given":"Yeye","family":"He","sequence":"additional","affiliation":[{"name":"Microsoft Research, Redmond, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0002-0071-1438","authenticated-orcid":false,"given":"Shihan","family":"Yu","sequence":"additional","affiliation":[{"name":"Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-4467-8170","authenticated-orcid":false,"given":"Qianqian","family":"Xu","sequence":"additional","affiliation":[{"name":"Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5625-2966","authenticated-orcid":false,"given":"Bin","family":"Wang","sequence":"additional","affiliation":[{"name":"Shanghai AI Lab, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1398-0621","authenticated-orcid":false,"given":"Guoliang","family":"Li","sequence":"additional","affiliation":[{"name":"Tsinghua University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8697-695X","authenticated-orcid":false,"given":"Conghui","family":"He","sequence":"additional","affiliation":[{"name":"Shanghai AI Lab, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0965-9058","authenticated-orcid":false,"given":"Fan","family":"Wu","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2026,5,18]]},"reference":[{"key":"e_1_2_1_1_1","unstructured":"[n.d.]. https:\/\/pdfdrive.webs.nf."},{"key":"e_1_2_1_2_1","unstructured":"[n.d.]. https:\/\/www.sci-hub.se."},{"key":"e_1_2_1_3_1","unstructured":"2025. DocAgent. https:\/\/github.com\/lisun-ai\/DocAgent"},{"key":"e_1_2_1_4_1","unstructured":"2025. m3docrag. https:\/\/github.com\/bloomberg\/m3docrag"},{"key":"e_1_2_1_5_1","unstructured":"2025. PyMuPDF 1.26.7. https:\/\/pypi.org\/project\/PyMuPDF\/"},{"key":"e_1_2_1_6_1","unstructured":"2025. SHTRAG. https:\/\/github.com\/Ruiying-Ma\/SHTRAG"},{"key":"e_1_2_1_7_1","unstructured":"2026. pdfplumber 0.11.9. https:\/\/pypi.org\/project\/pdfplumber\/"},{"key":"e_1_2_1_8_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00103"},{"key":"e_1_2_1_9_1","doi-asserted-by":"publisher","DOI":"10.14778\/3626292.3626294"},{"key":"e_1_2_1_10_1","volume-title":"sdf","author":"Bai Yushi","year":"2025","unstructured":"Yushi Bai, Shangqing Tu, Jiajie Zhang, and et alsdf. 2025. LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks. In ACL (1). Association for Computational Linguistics, 3639-3664."},{"key":"e_1_2_1_11_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00439"},{"key":"e_1_2_1_12_1","volume-title":"SV-RAG: LoRA-Contextualizing Adaptation of MLLMs for Long Document Understanding. arXiv preprint arXiv:2411.01106","author":"Chen Jian","year":"2024","unstructured":"Jian Chen, Ruiyi Zhang, Yufan Zhou, Tong Yu, Franck Dernoncourt, Jiuxiang Gu, Ryan A Rossi, Changyou Chen, and Tong Sun. 2024. SV-RAG: LoRA-Contextualizing Adaptation of MLLMs for Long Document Understanding. arXiv preprint arXiv:2411.01106 (2024)."},{"key":"e_1_2_1_13_1","volume-title":"M3docrag: Multi-modal retrieval is what you need for multi-page multi-document understanding. arXiv preprint arXiv:2411.04952","author":"Cho Jaemin","year":"2024","unstructured":"Jaemin Cho, Debanjan Mahata, Ozan Irsoy, Yujie He, and Mohit Bansal. 2024. M3docrag: Multi-modal retrieval is what you need for multi-page multi-document understanding. arXiv preprint arXiv:2411.04952 (2024)."},{"key":"e_1_2_1_14_1","unstructured":"Cheng Cui Ting Sun Manhui Lin Tingquan Gao Yubo Zhang Jiaxuan Liu Xueqing Wang Zelun Zhang Changda Zhou Hongen Liu Yue Zhang Wenyu Lv Kui Huang Yichao Zhang Jing Zhang Jun Zhang Yi Liu Dianhai Yu and Yanjun Ma. 2025. PaddleOCR 3.0 Technical Report. arXiv:2507.05595 [cs.CV] https:\/\/arxiv.org\/abs\/2507.05595"},{"key":"e_1_2_1_15_1","unstructured":"Manuel Faysse Hugues Sibille Tony Wu Bilel Omrani Gautier Viaud C\u00e9line Hudelot and Pierre Colombo. 2025. ColPali: Efficient Document Retrieval with Vision Language Models. arXiv:2407.01449 [cs.IR] https:\/\/arxiv.org\/abs\/2407.01449"},{"key":"e_1_2_1_16_1","volume-title":"Analyzing the efficacy of an llm-only approach for image-based document question answering. arXiv preprint arXiv:2309.14389","author":"Hegde Nidhi","year":"2023","unstructured":"Nidhi Hegde, Sujoy Paul, Gagan Madan, and Gaurav Aggarwal. 2023. Analyzing the efficacy of an llm-only approach for image-based document question answering. arXiv preprint arXiv:2309.14389 (2023)."},{"key":"e_1_2_1_17_1","volume-title":"mplugdocowl2: High-resolution compressing for ocr-free multi-page document understanding. arXiv preprint arXiv:2409.03420","author":"Hu Anwen","year":"2024","unstructured":"Anwen Hu, Haiyang Xu, Liang Zhang, Jiabo Ye, Ming Yan, Ji Zhang, Qin Jin, Fei Huang, and Jingren Zhou. 2024. mplugdocowl2: High-resolution compressing for ocr-free multi-page document understanding. arXiv preprint arXiv:2409.03420 (2024)."},{"key":"e_1_2_1_18_1","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3548112"},{"key":"e_1_2_1_19_1","volume-title":"Parameswaran","author":"Lin Yiming","year":"2025","unstructured":"Yiming Lin, Mawil Hasan, Rohan Kosalge, Alvin Cheung, and Aditya G. Parameswaran. 2025. TWIX: Automatically Reconstructing Structured Data from Templatized Documents. CoRR abs\/2501.06659 (2025)."},{"key":"e_1_2_1_20_1","volume-title":"Towards accurate and efficient document analytics with large language models. arXiv preprint arXiv:2405.04674","author":"Lin Yiming","year":"2024","unstructured":"Yiming Lin, Madelon Hulsebos, Ruiying Ma, Shreya Shankar, Sepanta Zeigham, Aditya G Parameswaran, and Eugene Wu. 2024. Towards accurate and efficient document analytics with large language models. arXiv preprint arXiv:2405.04674 (2024)."},{"key":"e_1_2_1_21_1","volume-title":"Proceedings of the Conference on Innovative Database Research (CIDR). 2.","author":"Liu Chunwei","year":"2025","unstructured":"Chunwei Liu, Matthew Russo, Michael Cafarella, Lei Cao, Peter Baile Chen, Zui Chen, Michael Franklin, Tim Kraska, Samuel Madden, Rana Shahout, et al. 2025. Palimpzest: Optimizing ai-powered analytics with declarative query processing. In Proceedings of the Conference on Innovative Database Research (CIDR). 2."},{"key":"e_1_2_1_22_1","unstructured":"OpenAI. 2025. Introducing GPT-5. Technical Report. https:\/\/openai.com\/index\/introducing-gpt-5\/"},{"key":"e_1_2_1_23_1","first-page":"1470","article-title":"Compositional Semantic Parsing on Semi-Structured Tables. In ACL (1)","author":"Pasupat Panupong","year":"2015","unstructured":"Panupong Pasupat and Percy Liang. 2015. Compositional Semantic Parsing on Semi-Structured Tables. In ACL (1). The Association for Computer Linguistics, 1470-1480.","journal-title":"The Association for Computer Linguistics"},{"key":"e_1_2_1_24_1","volume-title":"Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track. 153-169","author":"Saad-Falcon Jon","year":"2024","unstructured":"Jon Saad-Falcon, Joe Barrow, Alexa Siu, Ani Nenkova, Seunghyun Yoon, Ryan A Rossi, and Franck Dernoncourt. 2024. Pdftriage: Question answering over long, structured documents. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing: Industry Track. 153-169."},{"key":"e_1_2_1_25_1","volume-title":"Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 17712-17727","author":"Sun Li","year":"2025","unstructured":"Li Sun, Liu He, Shuyue Jia, Yangfan He, and Chenyu You. 2025. Docagent: An agentic framework for multi-modal long-context document understanding. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 17712-17727."},{"key":"e_1_2_1_26_1","doi-asserted-by":"publisher","DOI":"10.14778\/3749646.3749713"},{"key":"e_1_2_1_27_1","volume-title":"Proc. ACM Manag. Data","author":"Tang Zirui","year":"2026","unstructured":"Zirui Tang, Boyu Niu, Xuanhe Zhou, Boxiu Li, Wei Zhou, Jiannan Wang, Guoliang Li, Xinyi Zhang, and Fan Wu. 2026. ST-Raptor: LLM-Powered Semi-Structured Table Question Answering. Proc. ACM Manag. Data (2026)."},{"key":"e_1_2_1_28_1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01845"},{"key":"e_1_2_1_29_1","unstructured":"Qwen Team. 2025. Qwen2.5-VL. https:\/\/qwenlm.github.io\/blog\/qwen2.5-vl\/"},{"key":"e_1_2_1_30_1","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2023.109834"},{"key":"e_1_2_1_31_1","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01789"},{"key":"e_1_2_1_32_1","volume-title":"Visual Transformers: Token-based Image Representation and Processing for Computer Vision. arXiv:2006.03677 [cs.CV]","author":"Wu Bichen","year":"2020","unstructured":"Bichen Wu, Chenfeng Xu, Xiaoliang Dai, Alvin Wan, Peizhao Zhang, Zhicheng Yan, Masayoshi Tomizuka, Joseph Gonzalez, Kurt Keutzer, and Peter Vajda. 2020. Visual Transformers: Token-based Image Representation and Processing for Computer Vision. arXiv:2006.03677 [cs.CV]"},{"key":"e_1_2_1_33_1","volume-title":"Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 3: Industry Track). 544-557","author":"Yang Chening","year":"2025","unstructured":"Chening Yang, Duy-Khanh Vu, Minh-Tien Nguyen, Xuan-Quang Nguyen, Linh Nguyen, and Hung Le. 2025. Superrag: Beyond rag with layout-aware graph modeling. In Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 3: Industry Track). 544-557."},{"key":"e_1_2_1_34_1","doi-asserted-by":"publisher","DOI":"10.1145\/3731445"},{"key":"e_1_2_1_35_1","volume-title":"Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models. arXiv preprint arXiv:2506.05176","author":"Zhang Yanzhao","year":"2025","unstructured":"Yanzhao Zhang, Mingxin Li, Dingkun Long, Xin Zhang, Huan Lin, Baosong Yang, Pengjun Xie, An Yang, Dayiheng Liu, Junyang Lin, Fei Huang, and Jingren Zhou. 2025. Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models. arXiv preprint arXiv:2506.05176 (2025)."}],"container-title":["Proceedings of the ACM on Management of Data"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3802089","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,18]],"date-time":"2026-05-18T18:33:02Z","timestamp":1779129182000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3802089"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5,18]]},"references-count":35,"journal-issue":{"issue":"3","published-print":{"date-parts":[[2026,5,18]]}},"alternative-id":["10.1145\/3802089"],"URL":"https:\/\/doi.org\/10.1145\/3802089","relation":{},"ISSN":["2836-6573"],"issn-type":[{"value":"2836-6573","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5,18]]}}}