M3DocDep is an LVLM pipeline that extracts multimodal block embeddings, scores parent-child edges with a biaffine head, decodes a valid dependency tree via MST, and produces section-path-annotated chunks, yielding reported gains of 28-39% on STEDS and 1-15% on nDCG/ANLS.
Lum- berchunker: Long-form narrative document segmen- tation
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
MultiDocFusion combines vision parsing, OCR, LLM-based hierarchy detection, and DFS grouping to produce structure-aware chunks that raise RAG retrieval precision by 8-15% and ANLS QA scores by 2-3% on industrial benchmarks.
Section and subsection chunking achieves highest recall on legal QA while complex methods like RAPTOR and contextual chunking perform worse and cost more.
citing papers explorer
-
M3DocDep: Multi-modal, Multi-page, Multi-document Dependency Chunking with Large Vision-Language Models
M3DocDep is an LVLM pipeline that extracts multimodal block embeddings, scores parent-child edges with a biaffine head, decodes a valid dependency tree via MST, and produces section-path-annotated chunks, yielding reported gains of 28-39% on STEDS and 1-15% on nDCG/ANLS.
-
MultiDocFusion: Hierarchical and Multimodal Chunking Pipeline for Enhanced RAG on Long Industrial Documents
MultiDocFusion combines vision parsing, OCR, LLM-based hierarchy detection, and DFS grouping to produce structure-aware chunks that raise RAG retrieval precision by 8-15% and ANLS QA scores by 2-3% on industrial benchmarks.
-
Chunking German Legal Code
Section and subsection chunking achieves highest recall on legal QA while complex methods like RAPTOR and contextual chunking perform worse and cost more.