LLMs are applied in a generative pipeline for extracting, normalizing, and interpreting eligibility criteria from securities prospectuses, achieving up to 91% precision in document-level decisions with a conservative bias.
hub
Donut: Document understanding transformer without OCR
14 Pith papers cite this work. Polarity classification is still indexing.
14
Pith papers citing it
hub tools
citation-role summary
background 1
baseline 1
citation-polarity summary
representative citing papers
Injecting pre-computed layout priors from RT-DETR into VLM prompts raises markdown F1 from 0.37 to 0.92 on a 10k-page OOD benchmark and cuts infinite-loop failures across domains.
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.
A model-agnostic Geometric Risk Controller reduces extreme errors in VLM-based OCR by requiring cross-view consensus before accepting outputs.
Multimodal LLMs process code as images to achieve up to 8x token compression, with visual cues like syntax highlighting aiding tasks and clone detection remaining resilient or even improving under compression.
Introduces OCR+PAGE-1 and OCR+PAGE-N prompting strategies that improve zero-shot multi-page handwritten document transcription by sharing context across pages.
Nougat applies a visual transformer to convert academic PDFs into markup language while accurately handling mathematical content on a new scientific document dataset.
Systematic ablation of TrOCR fine-tuning for medieval HTR finds that freezing up to three encoder or six decoder layers does not significantly harm accuracy and that removing CLAHE contrast normalization yields comparable 7.84% CER on the Cortonese manuscript.
FastOCR dynamically selects a small subset of visual tokens per decoding step using focal-guided pruning and cross-step reuse, retaining 98% accuracy on Qwen2.5-VL while attending to only 5% of tokens and cutting attention latency by 3x.
MADP multi-agent pipeline with human-in-the-loop achieves 97% full automation on 955 real documents, 98.5% accuracy on ablation set, and 69-70% reductions in FTE, energy, and emissions versus manual processing.
Frontier multimodal LLMs achieve ~85% accuracy and ~90% weighted F1 on digitizing complex handwritten medical forms, with Gemini 3.1 strongest overall and prompt optimization lifting macro metrics over 60%.
MolSeek-OCR reaches exact SMILES matching accuracy comparable to leading image-to-sequence OCSR models after two-stage fine-tuning on PubChem renderings and USPTO-MOL patent images, but remains below image-to-graph state-of-the-art.
Fine-tuned multilingual LLMs achieve top shared-task scores on financial causality extraction in English and Spanish.
Specialized multimodal transformers outperform LLM-based models on visually rich documents, with image information contributing most to classification accuracy.
citing papers explorer
-
LLM-Based Examination of Eligibility Criteria from Securities Prospectuses at the German Central Bank
LLMs are applied in a generative pipeline for extracting, normalizing, and interpreting eligibility criteria from securities prospectuses, achieving up to 91% precision in document-level decisions with a conservative bias.
-
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.
-
From Plausibility to Verifiability: Risk-Controlled Generative OCR with Vision-Language Models
A model-agnostic Geometric Risk Controller reduces extreme errors in VLM-based OCR by requiring cross-view consensus before accepting outputs.
-
CodeOCR: On the Effectiveness of Vision Language Models in Code Understanding
Multimodal LLMs process code as images to achieve up to 8x token compression, with visual cues like syntax highlighting aiding tasks and clone detection remaining resilient or even improving under compression.
-
Judge a Book by its Cover: Investigating Multi-Modal LLMs for Multi-Page Handwritten Document Transcription
Introduces OCR+PAGE-1 and OCR+PAGE-N prompting strategies that improve zero-shot multi-page handwritten document transcription by sharing context across pages.
-
TrOCR for Medieval HTR: A Systematic Ablation Study with Cross-Dataset Validation
Systematic ablation of TrOCR fine-tuning for medieval HTR finds that freezing up to three encoder or six decoder layers does not significantly harm accuracy and that removing CLAHE contrast normalization yields comparable 7.84% CER on the Cortonese manuscript.
-
FastOCR: Dynamic Visual Fixation via KV Cache Pruning for Efficient Document Parsing
FastOCR dynamically selects a small subset of visual tokens per decoding step using focal-guided pruning and cross-step reuse, retaining 98% accuracy on Qwen2.5-VL while attending to only 5% of tokens and cutting attention latency by 3x.
-
From Handwriting to Structured Data: Benchmarking AI Digitisation of Handwritten Forms
Frontier multimodal LLMs achieve ~85% accuracy and ~90% weighted F1 on digitizing complex handwritten medical forms, with Gemini 3.1 strongest overall and prompt optimization lifting macro metrics over 60%.
-
Fine-tuning DeepSeek-OCR-2 for Molecular Structure Recognition
MolSeek-OCR reaches exact SMILES matching accuracy comparable to leading image-to-sequence OCSR models after two-stage fine-tuning on PubChem renderings and USPTO-MOL patent images, but remains below image-to-graph state-of-the-art.
-
Causal Connections: Leveraging Multilingual Fine-Tuning for Financial QA@FinCausal 2026
Fine-tuned multilingual LLMs achieve top shared-task scores on financial causality extraction in English and Spanish.
-
Multimodal Approaches for Visually-Rich Document Type Classification: A Comparative Analysis
Specialized multimodal transformers outperform LLM-based models on visually rich documents, with image information contributing most to classification accuracy.