mlsanityData to 14 Oct 2026

Question-Based Retrieval using Atomic Units for Enterprise RAG

arXiv 2405.12363 ↗First posted 20 May 2024cs.CLcited by 17

On the map: Retrieval-augmented generation → RAG pipeline design and evaluation

Abstract (arXiv)

Enterprise retrieval augmented generation (RAG) offers a highly flexible framework for combining powerful large language models (LLMs) with internal, possibly temporally changing, documents. In RAG, documents are first chunked. Relevant chunks are then retrieved for a user query, which are passed as context to a synthesizer LLM to generate the query response. However, the retrieval step can limit performance, as incorrect chunks can lead the synthesizer LLM to generate a false response. This work applies a zero-shot adaptation of standard dense retrieval steps for more accurate chunk recall. Specifically, a chunk is first decomposed into atomic statements. A set of synthetic questions are then generated on these atoms (with the chunk as the context). Dense retrieval involves finding the closest set of synthetic questions, and associated chunks, to the user query. It is found that retrieval with the atoms leads to higher recall than retrieval with chunks. Further performance gain is observed with retrieval using the synthetic questions generated over the atoms. Higher recall at the retrieval step enables higher performance of the enterprise LLM using the RAG pipeline.

Key facts. Cited by 17 papers in the map (data through 14 Oct 2026). Placed in RAG pipeline design and evaluation, one of 15 directions in Retrieval-augmented generation. Builds on 34 of its 56 references that are in the map. First posted on arXiv 20 May 2024.

Builds on

34 of 56 references.

Closest ancestors · 2

Same direction, newest first

  1. 2024Retrieval-Augmented Generation for AI-Generated Content: A Survey29 Feb
  2. 2023Retrieval-Augmented Generation for Large Language Models: A Survey18 Dec

Other references · 32

  1. 2017Attention Is All You Needcited by 60,361
  2. 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 13,028
  3. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  4. 2019Sentence-BERT: Sentence Embeddings using Siamese BERT-Networkscited by 9,170
  5. 2016SQuAD: 100,000+ Questions for Machine Comprehension of Textcited by 4,148
  6. 2021SimCSE: Simple Contrastive Learning of Sentence Embeddingscited by 2,173
  7. 2018Know What You Don't Know: Unanswerable Questions for SQuADcited by 1,554
  8. 2023FActScore: Fine-grained Atomic Evaluation of Factual Precision in Long Form Text Generationcited by 1,282
  9. 2022Text Embeddings by Weakly-Supervised Contrastive Pre-trainingcited by 1,253
  10. 2021BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Modelscited by 1,252
  11. 2023Towards General Text Embeddings with Multi-stage Contrastive Learningcited by 723
  12. 2023An Empirical Study of Catastrophic Forgetting in Large Language Models During Continual Fine-tuningcited by 639
20 more
  1. 2024RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrievalcited by 532
  2. 2022Precise Zero-Shot Dense Retrieval without Relevance Labelscited by 517
  3. 2021Sentence-T5: Scalable Sentence Encoders from Pre-trained Text-to-Text Modelscited by 463
  4. 2024LLM2Vec: Large Language Models Are Secretly Powerful Text Encoderscited by 404
  5. 2023Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergycited by 346
  6. 2023Large Language Models are Effective Text Rankers with Pairwise Ranking Promptingcited by 307
  7. 2024RAFT: Adapting Language Model to Domain Specific RAGcited by 262
  8. 2024Generative Representational Instruction Tuningcited by 233
  9. 2021ConSERT: A Contrastive Framework for Self-Supervised Sentence Representation Transfercited by 231
  10. 2023RA-DIT: Retrieval-Augmented Dual Instruction Tuningcited by 195
  11. 2023Chain-of-Note: Enhancing Robustness in Retrieval-Augmented Language Modelscited by 184
  12. 2022DiffCSE: Difference-based Contrastive Learning for Sentence Embeddingscited by 132
  13. 2023Information Retrieval: Recent Advances and Beyondcited by 49
  14. 2022Multi-View Document Representation Learning for Open-Domain Dense Retrievalcited by 41
  15. 2023Large Language Models are Strong Zero-Shot Retrievercited by 26
  16. 2024CLAPNQ: Cohesive Long-form Answers from Passages in Natural Questions for RAG systemscited by 25
  17. 2023GAR-meets-RAG Paradigm for Zero-Shot Information Retrievalcited by 18
  18. 2019BiPaR: A Bilingual Parallel Dataset for Multilingual and Cross-lingual Reading Comprehension on Novelscited by 17
  19. 2024Re3val: Reinforced and Reranked Generative Retrievalcited by 9
  20. 2021Fine-tune the Entire RAG Architecture (including DPR retriever) for Question-Answeringcited by 7

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2024: 3; 2025: 10; 2026: 4.

17 papers.

Sort:

By area:

  1. 2024Retrieval Augmented Generation (RAG) and Beyond: A Comprehensive Survey on How to Make your LLMs use External Data More Wiselycited by 66
  2. 2025MMDocIR: Benchmarking Multimodal Retrieval for Long Documentscited by 43
  3. 2024Ragnar"ok: A Reusable RAG Framework and Baselines for TREC 2024 Retrieval-Augmented Generation Trackcited by 32
  4. 2025FreshStack: Building Realistic Benchmarks for Evaluating Retrieval on Technical Documentscited by 15
  5. 2025ATOM: AdapTive and OptiMized dynamic temporal knowledge graph construction using LLMscited by 7
  6. 2025HeteRAG: A Heterogeneous Retrieval-augmented Generation Framework with Decoupled Knowledge Representationscited by 6
  7. 2025The Next Phase of Scientific Fact-Checking: Advanced Evidence Retrieval from Complex Structured Academic Paperscited by 6
  8. 2025HASH-RAG: Bridging Deep Hashing with Retriever for Efficient, Fine Retrieval and Augmented Generationcited by 5
  9. 2025REIS: A High-Performance and Energy-Efficient Retrieval System with In-Storage Processingcited by 4
  10. 2025Metadata-Driven Retrieval-Augmented Generation for Financial Question Answeringcited by 4
  11. 2025Fact-checking AI-generated news reports: Can LLMs catch their own lies?cited by 2
  12. 2026Evaluating Chunking Strategies For Retrieval-Augmented Generation in Oil and Gas Enterprise Documentscited by 1
5 more
  1. 2024The Geometry of Queries: Query-Based Innovations in Retrieval-Augmented Generation for Healthcare QAcited by 0
  2. 2025A Lightweight Multi-Expert Generative Language Model System for Engineering Information and Knowledge Extractioncited by 0
  3. 2026From Naive RAG to Deep Agentic Retrieval: An Evolving Context Engineering Pipeline for Regulatory Compliancecited by 0
  4. 2026VikingRAG: Accurate and Token-efficient Retrieval-augmented Generation over Structured Documentscited by 0
  5. 2026Co-adaptive Rule Routing and Constraining for Structured Knowledge Retrievalcited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2025A Systematic Review of Key Retrieval-Augmented Generation (RAG) Systems: Progress, Gaps, and Future Directions
  2. 2024Meta Knowledge for Retrieval Augmented Large Language Models
  3. 2024Improving Retrieval for RAG based Question Answering Models on Financial Documents
  4. 2024Optimizing and Evaluating Enterprise Retrieval-Augmented Generation (RAG): A Content Design Perspective
  5. 2024Toward Optimal Search and Retrieval for RAG
  6. 2024Introducing a new hyper-parameter for RAG: Context Window Utilization
  7. 2025Cache-Craft: Managing Chunk-Caches for Efficient Retrieval-Augmented Generation
  8. 2025Multiple Abstraction Level Retrieve Augment Generation