mlsanityData to 14 Oct 2026

Fishing for Answers: Exploring One-shot vs. Iterative Retrieval Strategies for Retrieval Augmented Generation

arXiv 2509.04820 ↗First posted 5 Sep 2025cs.IRcited by 2

On the map: Retrieval-augmented generation → Legal and financial RAG

Abstract (arXiv)

Retrieval-Augmented Generation (RAG) based on Large Language Models (LLMs) is a powerful solution to understand and query the industry's closed-source documents. However, basic RAG often struggles with complex QA tasks in legal and regulatory domains, particularly when dealing with numerous government documents. The top-k strategy frequently misses golden chunks, leading to incomplete or inaccurate answers. To address these retrieval bottlenecks, we explore two strategies to improve evidence coverage and answer quality. The first is a One-SHOT retrieval method that adaptively selects chunks based on a token budget, allowing as much relevant content as possible to be included within the model's context window. Additionally, we design modules to further filter and refine the chunks. The second is an iterative retrieval strategy built on a Reasoning Agentic RAG framework, where a reasoning LLM dynamically issues search queries, evaluates retrieved results, and progressively refines the context over multiple turns. We identify query drift and retrieval laziness issues and further design two modules to tackle them. Through extensive experiments on a dataset of government documents, we aim to offer practical insights and guidance for real-world applications in legal and regulatory domains.

Key facts. Cited by 2 papers in the map (data through 14 Oct 2026). Placed in Legal and financial RAG, one of 15 directions in Retrieval-augmented generation. Builds on 23 of its 41 references that are in the map. First posted on arXiv 5 Sep 2025.

Builds on

23 of 41 references.

References · 23

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2025Qwen3 Technical Reportcited by 8,820
  3. 2022ReAct: Synergizing Reasoning and Acting in Language Modelscited by 8,695
  4. 2023A Survey of Large Language Modelscited by 2,576
  5. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,212
  6. 2023A Survey on Hallucination in Large Language Models: Principles, Taxonomy, Challenges, and Open Questionscited by 1,601
  7. 2025Search-R1: Training LLMs to Reason and Leverage Search Engines with Reinforcement Learningcited by 1,444
  8. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,075
  9. 2023Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Modelscited by 698
  10. 2025Search-o1: Agentic Search-Enhanced Large Reasoning Modelscited by 433
  11. 2024Retrieval-Augmented Generation for AI-Generated Content: A Surveycited by 354
  12. 2023Benchmarking Large Language Models in Retrieval-Augmented Generationcited by 312
11 more
  1. 2025DeepResearcher: Scaling Deep Research via Reinforcement Learning in Real-world Environmentscited by 286
  2. 2023RA-DIT: Retrieval-Augmented Dual Instruction Tuningcited by 195
  3. 2023The Troubling Emergence of Hallucination in Large Language Models -- An Extensive Definition, Quantification, and Prescriptive Remediationscited by 131
  4. 2024Enhancing LLM Factual Accuracy with RAG to Counter Hallucinations: A Case Study on Domain-Specific Queries in Private Knowledge-Basescited by 77
  5. 2025DeepRetrieval: Hacking Real Search Engines and Retrievers with Large Language Models via Reinforcement Learningcited by 66
  6. 2023Controlling Large Language Model-based Agents for Large-Scale Decision-Making: An Actor-Critic Approachcited by 38
  7. 2025PENCIL: Long Thoughts with Short Memorycited by 33
  8. 2024Are Large Language Models Good Statisticians?cited by 27
  9. 2025Reasoning RAG via System 1 or System 2: A Survey on Reasoning Agentic Retrieval-Augmented Generation for Industry Challengescited by 19
  10. 2024Agentic Retrieval-Augmented Generation for Time Series Analysiscited by 14
  11. 2025ReZero: Enhancing LLM search ability by trying one-more-timecited by 9

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026SEEK: Steering LLM Reasoning for RAG via Internal Reasoning Sketchescited by 2
  2. 2026G2-Reader: Dual Evolving Graphs for Multimodal Document QAcited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2026Stateful Evidence-Driven Retrieval-Augmented Generation with Iterative Reasoning
  2. 2025All for law and law for all: Adaptive RAG Pipeline for Legal Research
  3. 2025On the Influence of Context Size and Model Choice in Retrieval-Augmented Generation Systems
  4. 2025A Survey on Knowledge-Oriented Retrieval-Augmented Generation
  5. 2025Retrieval-Augmented Generation: A Comprehensive Survey of Architectures, Enhancements, and Robustness Frontiers
  6. 2025Evaluation of retrieval-based QA on QUEST-LOFT
  7. 2026Mapping the RAG Landscape: A Four Axis Taxonomy of Efficiency, Defense, Interactivity, and Reasoning
  8. 2025Stronger Baselines for Retrieval-Augmented Generation with Long-Context Language Models