Adding overlap between chunks: use num_overlap_characters
Unstructured - Feature request: in addition to num_characters and max_characters in the chunking strategy parameter, add num_overlap_characters.
## The fix
[scanny (commenter)]: Chunking now supports an overlap parameter: pass num_overlap_characters alongside your chunking strategy to keep overlapping text between consecutive chunks. This keeps semantic continuity across chunk boundaries.
Thread: gh:Unstructured-IO/unstructured#1706