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.
MinerU2.5-Pro: Pushing the Limits of Data-Centric Document Parsing at Scale
7 Pith papers cite this work. Polarity classification is still indexing.
7
Pith papers citing it
abstract
Current document parsing methods advance primarily through model architecture innovation, while systematic engineering of training data remains underexplored. Yet state-of-the-art models spanning diverse architectures and parameter scales exhibit highly consistent failure patterns on the same set of hard samples, suggesting that the performance bottleneck stems from shared deficiencies in training data rather than from architectural differences. Building on this finding, we present MinerU2.5-Pro, which advances the state of the art purely through data engineering and training strategy design while retaining the 1.2B-parameter architecture of MinerU2.5 unchanged. At its core is a Data Engine co-designed around coverage, informativeness, and annotation accuracy: Diversity-and-Difficulty-Aware Sampling expands training data from under 10M to 65.5M samples while mitigating distribution shift; Cross-Model Consistency Verification leverages output consensus among heterogeneous models to assess sample difficulty and generate reliable annotations; the Judge-and-Refine pipeline improves annotation quality for hard samples through render-then-verify iterative correction. A three-stage progressive training strategy--large-scale pre-training, hard sample fine-tuning, and GRPO alignment--sequentially exploits these data at different quality tiers. On the evaluation front, we rectify element-matching biases in OmniDocBench v1.5 and introduce a Hard subset, establishing the more discriminative OmniDocBench v1.6 protocol. Without any architectural modification, MinerU2.5-Pro achieves 95.69 on OmniDocBench v1.6, improving over the same-architecture baseline by 2.71 points and surpassing all existing methods, including those based on models with over 200x more parameters.
citation-role summary
baseline 1
citation-polarity summary
years
2026 7roles
baseline 1polarities
baseline 1representative 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.
MedSP1000 benchmark shows top LLMs complete at most 60.4% of expert rubric items during multi-turn standardized patient simulations.
Dr. DocBench is a difficulty-aware benchmark of 4,514 pages and 65k annotations from multilingual books, demonstrating that strong results on prior document parsing tests do not transfer to expert-level cases.
CMIP-Forge presents a retrieval-augmented agentic system with automated guardrails and adversarial self-review for autonomous execution of climate research tasks on CMIP6 literature and ESGF data.
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.
ParseFixer combines full-page backbone parsing with agentic selective multimodal correction to reach third place (score 61.78) in the DataMFM Challenge Track 1.
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.
-
Evaluating Large Language Models in Dynamic Clinical Decision-Making with Standardized Patient Cases
MedSP1000 benchmark shows top LLMs complete at most 60.4% of expert rubric items during multi-turn standardized patient simulations.
-
Dr. DocBench: A Comprehensive Benchmark for Expert-Level and Difficult Document Parsing
Dr. DocBench is a difficulty-aware benchmark of 4,514 pages and 65k annotations from multilingual books, demonstrating that strong results on prior document parsing tests do not transfer to expert-level cases.
-
CMIP-Forge: An Agentic System that Retrieves, Computes, and Self-Reviews Climate Science
CMIP-Forge presents a retrieval-augmented agentic system with automated guardrails and adversarial self-review for autonomous execution of climate research tasks on CMIP6 literature and ESGF data.
-
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.
-
ParseFixer: An Agentic Framework for Document Parsing via Selective Multimodal Correction
ParseFixer combines full-page backbone parsing with agentic selective multimodal correction to reach third place (score 61.78) in the DataMFM Challenge Track 1.