CiteVQA requires models to cite specific document regions with bounding boxes alongside answers and finds that even the strongest MLLMs frequently cite the wrong region, with top SAA scores of only 76.0 for closed models and 22.5 for open-source ones.
hub Canonical reference
MinerU2.5: A Decoupled Vision-Language Model for Efficient High-Resolution Document Parsing
Canonical reference. 71% of citing Pith papers cite this work as background.
26
Pith papers citing it
Background
71% of classified citations
abstract
We introduce MinerU2.5, a 1.2B-parameter document parsing vision-language model that achieves state-of-the-art recognition accuracy while maintaining exceptional computational efficiency. Our approach employs a coarse-to-fine, two-stage parsing strategy that decouples global layout analysis from local content recognition. In the first stage, the model performs efficient layout analysis on downsampled images to identify structural elements, circumventing the computational overhead of processing high-resolution inputs. In the second stage, guided by the global layout, it performs targeted content recognition on native-resolution crops extracted from the original image, preserving fine-grained details in dense text, complex formulas, and tables. To support this strategy, we developed a comprehensive data engine that generates diverse, large-scale training corpora for both pretraining and fine-tuning. Ultimately, MinerU2.5 demonstrates strong document parsing ability, achieving state-of-the-art performance on multiple benchmarks, surpassing both general-purpose and domain-specific models across various recognition tasks, while maintaining significantly lower computational overhead.
hub tools
citation-role summary
background 5
baseline 2
citation-polarity summary
representative citing papers
PureDocBench shows document parsing is far from solved, with top models at ~74/100, small specialists competing with large VLMs, and ranking reversals under real degradation.
The paper presents UniPPTBench and UniPPTEval, a unified benchmark and scenario-aware evaluation framework for presentation generation from vague prompts, long documents, multimodal documents, and multi-source inputs.
Introduces a feature-level annotated patent dataset and LLM retrieval-reasoning workflows that outperform embedding baselines on passage retrieval and novel feature identification while avoiding spurious correlations in novelty prediction.
ColChunk adaptively chunks visual document patches into contextual multi-vectors via clustering, cutting storage by over 90% while raising average nDCG@5 by 9 points.
A fixed 1.2B model trained via diversity-aware sampling, cross-model verification, annotation refinement, and progressive stages achieves new state-of-the-art document parsing accuracy of 95.69 on OmniDocBench v1.6.
Q-Mask uses query-conditioned causal masks to separate text location from recognition in OCR VLMs, backed by a new benchmark and 26M-pair training dataset.
Visual-ERM is a new multimodal reward model that supplies fine-grained visual feedback for training vision-language models on chart-to-code, table, and SVG tasks, yielding measurable gains over prior rewards.
FinCriticalED benchmark reveals that OCR and MLLM systems frequently fail to preserve critical financial facts such as numbers and monetary units even when lexical accuracy is high.
An end-to-end document parser trained on 5M synthetic and mined pages with multi-task RL reaches 87.6% on olmOCR-Bench and 74.3% on ParseBench.
StrucTab achieves SOTA table parsing performance by unifying structural subtasks through sequential reasoning and using decomposed RL rewards in Uni-TabRL, plus a new TableVerse-5K benchmark.
MPDocBench-Parse provides 433 annotated multi-page documents and an evaluation protocol covering text/table/formula extraction, merging, figure extraction, reading order, and heading hierarchy for realistic document parsing.
JSON schema constraints improve LLM extraction of nested quantum cascade laser structures to 83.4% F1, delivering up to 24.1% gains for smaller models.
VLMs suffer from a perceptual bandwidth bottleneck; the paper formalizes active visual reasoning as sequential Bayesian optimal experimental design, derives a coverage-resolution proxy objective, and introduces the training-free FOVEA method that yields gains on high-resolution benchmarks.
InstructTable combines instruction-guided pre-training on structural patterns with visual fine-tuning and a template-free synthetic data generator (TME) to reach state-of-the-art table structure recognition on public benchmarks and a new complex-table test set.
A parser-oriented refinement stage performs set-level reasoning on detector hypotheses to jointly decide instance retention, refine boxes, and set parser input order, cutting reading order errors to 0.024 on OmniDocBench.
PaddleOCR-VL uses a Valid Region Focus Module to select key visual tokens and a 0.9B model for guided recognition, delivering SOTA document parsing with far fewer tokens and parameters.
Omni Parsing framework converts complex multimodal signals into locatable, enumerable, and traceable structured knowledge via hierarchical detection, recognition, and interpreting with strict evidence alignment.
ChartVerse uses Rollout Posterior Entropy and truth-anchored inverse QA synthesis to produce 640K high-quality chart reasoning samples, training an 8B model that surpasses its 30B teacher.
Presents RT-DocLayout, a 33M-parameter end-to-end model extending RT-DETR that unifies layout classification, detection, segmentation, and reading-order prediction at 132.1 FPS with claimed SOTA results on public benchmarks.
ToolGate is a pre-call controller that reduces token usage in tool-augmented VLM agents to 64-69% of baseline while preserving or slightly improving accuracy on benchmarks.
ABot-OCR is a new end-to-end VLM for direct image-to-Markdown transcription using a custom data engine and structure-constrained RL optimization, reporting SOTA scores of 92.81/93.30 on OmniDocBench v1.5/v1.6.
AiraXiv is a proposed AI-driven platform for open preprints that supports human and AI authors with interactive UI and MCP-based interactions, validated by serving as the submission system for ICAIS 2025.
PaddleOCR-VL-1.5 is a 0.9B VLM achieving 94.5% SOTA accuracy on OmniDocBench v1.5, with added robustness to physical distortions and support for seal recognition plus text spotting.
citing papers explorer
-
CiteVQA: Benchmarking Evidence Attribution for Trustworthy Document Intelligence
CiteVQA requires models to cite specific document regions with bounding boxes alongside answers and finds that even the strongest MLLMs frequently cite the wrong region, with top SAA scores of only 76.0 for closed models and 22.5 for open-source ones.
-
How Far Is Document Parsing from Solved? PureDocBench: A Source-TraceableBenchmark across Clean, Degraded, and Real-World Settings
PureDocBench shows document parsing is far from solved, with top models at ~74/100, small specialists competing with large VLMs, and ranking reversals under real degradation.
-
UniPPTBench: A Unified Benchmark for Presentation Generation Across Diverse Input Settings
The paper presents UniPPTBench and UniPPTEval, a unified benchmark and scenario-aware evaluation framework for presentation generation from vague prompts, long documents, multimodal documents, and multi-source inputs.
-
Is It Novel and Why? Fine-Grained Patent Novelty Prediction Based on Passage Retrieval
Introduces a feature-level annotated patent dataset and LLM retrieval-reasoning workflows that outperform embedding baselines on passage retrieval and novel feature identification while avoiding spurious correlations in novelty prediction.
-
Visual Late Chunking: An Empirical Study of Contextual Chunking for Efficient Visual Document Retrieval
ColChunk adaptively chunks visual document patches into contextual multi-vectors via clustering, cutting storage by over 90% while raising average nDCG@5 by 9 points.
-
MinerU2.5-Pro: Pushing the Limits of Data-Centric Document Parsing at Scale
A fixed 1.2B model trained via diversity-aware sampling, cross-model verification, annotation refinement, and progressive stages achieves new state-of-the-art document parsing accuracy of 95.69 on OmniDocBench v1.6.
-
Q-Mask: Query-driven Causal Masks for Text Anchoring in OCR-Oriented Vision-Language Models
Q-Mask uses query-conditioned causal masks to separate text location from recognition in OCR VLMs, backed by a new benchmark and 26M-pair training dataset.
-
Visual-ERM: Reward Modeling for Visual Equivalence
Visual-ERM is a new multimodal reward model that supplies fine-grained visual feedback for training vision-language models on chart-to-code, table, and SVG tasks, yielding measurable gains over prior rewards.
-
FinCriticalED: A Visual Benchmark for Financial Fact-Level OCR
FinCriticalED benchmark reveals that OCR and MLLM systems frequently fail to preserve critical financial facts such as numbers and monetary units even when lexical accuracy is high.
-
Infinity-Parser2 Technical Report
An end-to-end document parser trained on 5M synthetic and mined pages with multi-task RL reaches 87.6% on olmOCR-Bench and 74.3% on ParseBench.
-
StrucTab: A Structured Optimization Framework for Table Parsing
StrucTab achieves SOTA table parsing performance by unifying structural subtasks through sequential reasoning and using decomposed RL rewards in Uni-TabRL, plus a new TableVerse-5K benchmark.
-
MPDocBench-Parse: Benchmarking Practical Multi-page Document Parsing
MPDocBench-Parse provides 433 annotated multi-page documents and an evaluation protocol covering text/table/formula extraction, merging, figure extraction, reading order, and heading hierarchy for realistic document parsing.
-
Information Extraction of Nested Complex Structure of Quantum Cascade Lasers via Large Language Models
JSON schema constraints improve LLM extraction of nested quantum cascade laser structures to 83.4% F1, delivering up to 24.1% gains for smaller models.
-
The Perceptual Bandwidth Bottleneck in Vision-Language Models: Active Visual Reasoning via Sequential Experimental Design
VLMs suffer from a perceptual bandwidth bottleneck; the paper formalizes active visual reasoning as sequential Bayesian optimal experimental design, derives a coverage-resolution proxy objective, and introduces the training-free FOVEA method that yields gains on high-resolution benchmarks.
-
InstructTable: Improving Table Structure Recognition Through Instructions
InstructTable combines instruction-guided pre-training on structural patterns with visual fine-tuning and a template-free synthetic data generator (TME) to reach state-of-the-art table structure recognition on public benchmarks and a new complex-table test set.
-
Parser-Oriented Structural Refinement for a Stable Layout Interface in Document Parsing
A parser-oriented refinement stage performs set-level reasoning on detector hypotheses to jointly decide instance retention, refine boxes, and set parser input order, cutting reading order errors to 0.024 on OmniDocBench.
-
Boosting Document Parsing Efficiency and Performance with Coarse-to-Fine Visual Processing
PaddleOCR-VL uses a Valid Region Focus Module to select key visual tokens and a 0.9B model for guided recognition, delivering SOTA document parsing with far fewer tokens and parameters.
-
Logics-Parsing-Omni Technical Report
Omni Parsing framework converts complex multimodal signals into locatable, enumerable, and traceable structured knowledge via hierarchical detection, recognition, and interpreting with strict evidence alignment.
-
ChartVerse: Scaling Chart Reasoning via Reliable Programmatic Synthesis from Scratch
ChartVerse uses Rollout Posterior Entropy and truth-anchored inverse QA synthesis to produce 640K high-quality chart reasoning samples, training an 8B model that surpasses its 30B teacher.
-
RT-DocLayout: Real-Time End-to-End Document Layout Analysis with Reading Order in the Wild
Presents RT-DocLayout, a 33M-parameter end-to-end model extending RT-DETR that unifies layout classification, detection, segmentation, and reading-order prediction at 132.1 FPS with claimed SOTA results on public benchmarks.
-
ToolGate: Token-Efficient Pre-Call Control for Tool-Augmented Vision-Language Agents
ToolGate is a pre-call controller that reduces token usage in tool-augmented VLM agents to 64-69% of baseline while preserving or slightly improving accuracy on benchmarks.
-
ABot-OCR Technical Report
ABot-OCR is a new end-to-end VLM for direct image-to-Markdown transcription using a custom data engine and structure-constrained RL optimization, reporting SOTA scores of 92.81/93.30 on OmniDocBench v1.5/v1.6.
-
AiraXiv: An AI-Driven Open-Access Platform for Human and AI Scientists
AiraXiv is a proposed AI-driven platform for open preprints that supports human and AI authors with interactive UI and MCP-based interactions, validated by serving as the submission system for ICAIS 2025.
-
PaddleOCR-VL-1.5: Towards a Multi-Task 0.9B VLM for Robust In-the-Wild Document Parsing
PaddleOCR-VL-1.5 is a 0.9B VLM achieving 94.5% SOTA accuracy on OmniDocBench v1.5, with added robustness to physical distortions and support for seal recognition plus text spotting.
-
PaddleOCR-VL-1.6: Expanding the Frontier of Document Parsing with Under-Optimized Region Refinement and Progressive Post-Training
PaddleOCR-VL-1.6 improves on PaddleOCR-VL-1.5 via region-aware data optimization and progressive post-training to reach 96.33% on OmniDocBench v1.6.
- CharTool: Tool-Integrated Visual Reasoning for Chart Understanding