Lune

ACL Findings2025

Dolphin: Document Image Parsing via Heterogeneous Anchor Prompting

Hao Feng, Shu Wei, Xiang Fei, Wei Shi, Yingdong Han, Lei Liao, Jinghui Lu, Binghong Wu, Qi Liu, Chunhui Lin, Jingqun Tang, Hao Liu, Can Huang

Abstract

Document image parsing is challenging due to its complexly intertwined elements such as text paragraphs, figures, formulas, and tables. Current approaches either assemble specialized expert models or directly generate pagelevel content autoregressively, facing integration overhead, efficiency bottlenecks, and layout structure degradation despite their decent performance. To address these limitations, we present Dolphin (Document Image Parsing via Heterogeneous Anchor Prompting), a novel multimodal document image parsing model following an analyze-then-parse paradigm. In the first stage, Dolphin generates a sequence of layout elements in reading order. These heterogeneous elements, serving as anchors and coupled with task-specific prompts, are fed back to Dolphin for parallel content parsing in the second stage. To train Dolphin, we construct a large-scale dataset of over 30 million samples, covering multi-granularity parsing tasks. Through comprehensive evaluations on both prevalent benchmarks and selfconstructed ones, Dolphin achieves state-ofthe-art performance across diverse page-level and element-level settings, while ensuring superior efficiency through its lightweight architecture and parallel parsing mechanism. The code and pre-trained models are publicly available at https://github.com/ByteDance/Dolphin

Ask Lune

Know this paper in minutes, not an afternoon.

Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.

15 top-tier papers have cited it since. Know what they found before your reviewers do.

Questions to start from

Lune can draw on

the full paper · the top-tier papers that cite it · related work

Get the answer

Free to start. No credit card required.

Lineage

  • LayoutLMv3: Pre-training for Document AI with Unified Text and Image Masking cites LayoutLM: Pre-training of Text and Layout for Document Image Understanding
Builds onCited byLeft to right by year; the more cited, the higher.

Cited by top-tier papers15

  1. DocSeeker: Structured Visual Reasoning with Evidence Grounding for Long Document UnderstandingHao Yan, Yuliang Liu, Xingchen Liu, Yuyi Zhang et al.CVPR 2026 · 9 citations
  2. Reading or Reasoning? Format Decoupled Reinforcement Learning for Document OCRYufeng Zhong, Lei Chen, Zhixiong Zeng, Xuanle Zhao et al.CVPR 2026 · 8 citations
  3. TRivia: Self-supervised Fine-tuning of Vision-Language Models for Table RecognitionJunyuan Zhang, Bin Wang, Qintong Zhang, Fan Wu et al.CVPR 2026 · 5 citations
  4. MORE: A Multilingual Document Parsing Benchmark and EvaluationLong Xu, Binghong Wu, TingHao YU, Hao Feng et al.ICML 2026 · 3 citations
  5. Boosting Document Parsing Efficiency and Performance with Coarse-to-Fine Visual ProcessingCheng Cui, Ting Sun, Suyin Liang, Tingquan Gao et al.CVPR 2026 · 3 citations

Builds on18

  1. Swin Transformer: Hierarchical Vision Transformer using Shifted WindowsZe Liu, Yutong Lin, Yue Cao, Han Hu et al.ICCV 2021 · 31,683 citations
  2. Grounding Multimodal Large Language Models to the WorldZhiliang Peng, Wenhui Wang, Li Dong, Yaru Hao et al.ICLR 2024 · 1,170 citations
  3. Efficient Memory Management for Large Language Model Serving with PagedAttentionWoosuk Kwon, Zhuohan Li, Siyuan Zhuang, Ying Sheng et al.SOSP 2023 · 1,016 citations
  4. LayoutLMv3: Pre-training for Document AI with Unified Text and Image MaskingYupan Huang, Tengchao Lv, Lei Cui, Yutong Lu et al.ACM MM 2022 · 606 citations
  5. LayoutLM: Pre-training of Text and Layout for Document Image UnderstandingYiheng Xu, Minghao Li, Lei Cui, Shaohan Huang et al.KDD 2020 · 575 citations

Related papers

  1. DocHieNet: A Large and Diverse Dataset for Document Hierarchy ParsingHangdi Xing, Changxu Cheng, Feiyu Gao, Zirui Shao et al.EMNLP 2024 · 3 citations
  2. Infinity-Parser: Layout-Aware Reinforcement Learning with High-quality Document Parsing DatasetBaode Wang, Biao Wu, Weizhen Li, Meng Fang et al.ACL Findings 2026 · 4 citations
  3. Towards Real-World Document Parsing via Realistic Scene Synthesis and Document-Aware TrainingGengluo Li, Pengyuan Lyu, Chengquan Zhang, Huawen Shen et al.CVPR 2026 · 9 citations
  4. Intelligent Document Parsing: Towards End-to-end Document Parsing via Decoupled Content Parsing and Layout GroundingHangdi Xing, Feiyu Gao, Qi Zheng, Zhaoqing Zhu et al.EMNLP Findings 2025
  5. SelfDoc: Self-Supervised Document Representation LearningPeizhao Li, Jiuxiang Gu, Jason Kuen, Vlad I. Morariu et al.CVPR 2021
Dusk over the sea between two cliffs drawn in fine vertical lines