mlsanityData to 14 Oct 2026

Passage Segmentation of Documents for Extractive Question Answering

arXiv 2501.09940 ↗First posted 17 Jan 2025cs.CL cs.IRcited by 6

On the map: Retrieval-augmented generation → Multimodal RAG

Abstract (arXiv)

Retrieval-Augmented Generation (RAG) has proven effective in open-domain question answering. However, the chunking process, which is essential to this pipeline, often receives insufficient attention relative to retrieval and synthesis components. This study emphasizes the critical role of chunking in improving the performance of both dense passage retrieval and the end-to-end RAG pipeline. We then introduce the Logits-Guided Multi-Granular Chunker (LGMGC), a novel framework that splits long documents into contextualized, self-contained chunks of varied granularity. Our experimental results, evaluated on two benchmark datasets, demonstrate that LGMGC not only improves the retrieval step but also outperforms existing chunking methods when integrated into a RAG pipeline.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Placed in Multimodal RAG, one of 15 directions in Retrieval-augmented generation. Builds on 17 of its 20 references that are in the map. First posted on arXiv 17 Jan 2025.

Builds on

17 of 20 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Grounding Language Model with Chunking-Free In-Context Retrieval15 Feb

Other references · 16

  1. 2024The Llama 3 Herd of Modelscited by 14,715
  2. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  3. 2020Dense Passage Retrieval for Open-Domain Question Answeringcited by 3,824
  4. 2023Self-RAG: Learning to Retrieve, Generate, and Critique through Self-Reflectioncited by 1,839
  5. 2023LongBench: A Bilingual, Multitask Benchmark for Long Context Understandingcited by 1,417
  6. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,253
  7. 2021Unsupervised Dense Information Retrieval with Contrastive Learningcited by 1,216
  8. 2017The NarrativeQA Reading Comprehension Challengecited by 597
  9. 2023C-Pack: Packed Resources For General Chinese Embeddingscited by 578
  10. 2023Ragas: Automated Evaluation of Retrieval Augmented Generationcited by 418
  11. 2023Improving Text Embeddings with Large Language Modelscited by 303
  12. 2024RAFT: Adapting Language Model to Domain Specific RAGcited by 262
4 more
  1. 2023Dense X Retrieval: What Retrieval Granularity Should We Use?cited by 88
  2. 2023Learning to Filter Context for Retrieval-Augmented Generationcited by 83
  3. 2024Lynx: An Open Source Hallucination Evaluation Modelcited by 57
  4. 2024LumberChunker: Long-Form Narrative Document Segmentationcited by 36

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 3; 2026: 3.

6 papers.

  1. 2025LightMem: Lightweight and Efficient Memory-Augmented Generationcited by 180
  2. 2025UniversalRAG: Retrieval-Augmented Generation over Corpora of Diverse Modalities and Granularitiescited by 20
  3. 2026A Systematic Investigation of Document Chunking Strategies and Embedding Sensitivitycited by 3
  4. 2026RAGTurk: Best Practices for Retrieval Augmented Generation in Turkishcited by 1
  5. 2025TextBite: A Historical Czech Document Dataset for Logical Page Segmentationcited by 0
  6. 2026SegNSP: Revisiting Next Sentence Prediction for Linear Text Segmentationcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2026Evaluating Chunking Strategies for Retrieval-Augmented Generation on Academic Texts
  2. 2025FreeChunker: A Cross-Granularity Chunking Framework
  3. 2026Comparing Chunking and Embedding Strategies for Turkish RAG Systems
  4. 2025MoC: Mixtures of Text Chunking Learners for Retrieval-Augmented Generation System
  5. 2024Is Semantic Chunking Worth the Computational Cost?
  6. 2025Enhancing Retrieval Augmented Generation with Hierarchical Text Segmentation Chunking
  7. 2025A New HOPE: Domain-agnostic Automatic Evaluation of Text Chunking
  8. 2024ChunkRAG: Novel LLM-Chunk Filtering Method for RAG Systems