From Retrieval to Recognition:How Vision--Language Models Become OCR Specialists

arXiv:2609.21543 · cs.CV · Submitted 2026-09-18 · Read on arXiv

cs.CV

Submitted: 2026-09-18

Updated: 2026-09-18

Terminology

• Ablation — 29× in this paper · explained in 1 episode
• Alignment — 7× in this paper · explained in 1 episode
• Core — 5× in this paper · explained in 2 episodes
• Calibration — 3× in this paper · explained in 7 episodes
• Commentary — 3× in this paper · explained in 2 episodes
• Grounding — 3× in this paper · explained in 3 episodes
• Accuracy — 2× in this paper · explained in 1 episode
• Interpretability — 2× in this paper · explained in 15 episodes
• Sparse Attention — 2× in this paper · explained in 5 episodes
• Trace — 2× in this paper · explained in 2 episodes
• Visual Grounding — 2× in this paper · explained in 2 episodes
• Adaptation — 1× in this paper · explained in 1 episode
• Consistency — 1× in this paper · explained in 3 episodes
• Fine-Tuning — 1× in this paper · explained in 5 episodes
• Hallucination — 1× in this paper · explained in 3 episodes
• In-Context Learning — 1× in this paper · explained in 4 episodes
• Sparsity — 1× in this paper · explained in 4 episodes
• Transformer — 1× in this paper · explained in 2 episodes
• alpha — 1× in this paper · explained in 1 episode

Sources

• Qwen3-VL Technical Report
• Nougat: Neural Optical Understanding for Academic Documents
• GLM-OCR Technical Report
• Can Retrieval Heads See Images? Multimodal Retrieval Heads in Long-Context Vision-Language Models
• Optimal ablation for interpretability
• In-context Learning and Induction Heads
• olmOCR: Unlocking Trillions of Tokens in PDFs with Vision Language Models
• Image Over Text: Transforming Formula Recognition Evaluation with Character Detection Matching
• Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution
• General OCR Theory: Towards OCR-2.0 via a Unified End-to-end Model
• DeepSeek-OCR: Contexts Optical Compression
• Retrieval Head Mechanistically Explains Long-Context Factuality
• PaddleOCR-VL-1.6: Expanding the Frontier of Document Parsing with Under-Optimized Region Refinement and Progressive Post-Training
• Image-based table recognition: data, model, and evaluation

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models