domains / families / research / vlm-ocr-pipeline

vlm-ocr-pipeline

separate

VLM-based OCR pipeline: model selection, prompts, architecture, evaluation.

When to use

• VLM-based OCR pipeline
• model selection
• prompts
• architecture
• evaluation
• scanned document corpora
• language-specific transcription prompt
• per-page JSON output schema

When not to use

• general VLM leaderboards
• MMMU
• VQAv2

Skill page

Source