An open-source document parsing model with SOTA accuracy and speed, using an RSI loop during post-training.
import re
from oxr.pipeline.models import Block, ContentFormat
from oxr.pipeline.postprocess.formula_postprocess import move_formula_numbering_outside_math
def postprocess_text(block: Block, raw_text: str) -> Block:
"""Clean up text block output."""
text = raw_text.strip()
# Remove some excessive whitespaces
text = re.sub(r'\n+', '\n', text)
text = move_formula_numbering_outside_math(text)
block.content = text
block.format = ContentFormat.TEXT
return block