mlsanityData to 14 Oct 2026

Mitigating Lost-in-Retrieval Problems in Retrieval Augmented Multi-Hop Question Answering

arXiv 2502.14245 ↗First posted 20 Feb 2025cs.CLcited by 22

On the map: Retrieval-augmented generation → Knowledge-graph and structured-knowledge RAG

Abstract (arXiv)

In this paper, we identify a critical problem, "lost-in-retrieval", in retrieval-augmented multi-hop question answering (QA): the key entities are missed in LLMs' sub-question decomposition. "Lost-in-retrieval" significantly degrades the retrieval performance, which disrupts the reasoning chain and leads to the incorrect answers. To resolve this problem, we propose a progressive retrieval and rewriting method, namely ChainRAG, which sequentially handles each sub-question by completing missing key entities and retrieving relevant sentences from a sentence graph for answer generation. Each step in our retrieval and rewriting process builds upon the previous one, creating a seamless chain that leads to accurate retrieval and answers. Finally, all retrieved sentences and sub-question answers are integrated to generate a comprehensive answer to the original question. We evaluate ChainRAG on three multi-hop QA datasets - MuSiQue, 2Wiki, and HotpotQA - using three large language models: GPT4o-mini, Qwen2.5-72B, and GLM-4-Plus. Empirical results demonstrate that ChainRAG consistently outperforms baselines in both effectiveness and efficiency.

Key facts. Cited by 22 papers in the map (data through 14 Oct 2026). Placed in Knowledge-graph and structured-knowledge RAG, one of 15 directions in Retrieval-augmented generation. Builds on 20 of its 36 references that are in the map. First posted on arXiv 20 Feb 2025.

Builds on

20 of 36 references.

Closest ancestors · 3

Same direction, newest first

  1. 2024From Local to Global: A Graph RAG Approach to Query-Focused Summarization24 Apr
  2. 2024RAPTOR: Recursive Abstractive Processing for Tree-Organized Retrieval31 Jan
  3. 2023Enhancing Retrieval-Augmented Large Language Models with Iterative Retrieval-Generation Synergy24 May

Other references · 17

  1. 2020Retrieval-Augmented Generation for Knowledge-Intensive NLP Taskscited by 9,297
  2. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,754
  3. 2023Retrieval-Augmented Generation for Large Language Models: A Surveycited by 2,212
  4. 2020DeBERTa: Decoding-enhanced BERT with Disentangled Attentioncited by 2,181
  5. 2023LongBench: A Bilingual, Multitask Benchmark for Long Context Understandingcited by 1,417
  6. 2020Constructing A Multi-hop QA Dataset for Comprehensive Evaluation of Reasoning Stepscited by 1,282
  7. 2024ChatGLM: A Family of Large Language Models from GLM-130B to GLM-4 All Toolscited by 1,121
  8. 2024M3-Embedding: Multi-Linguality, Multi-Functionality, Multi-Granularity Text Embeddings Through Self-Knowledge Distillationcited by 1,073
  9. 2022Interleaving Retrieval with Chain-of-Thought Reasoning for Knowledge-Intensive Multi-Step Questionscited by 923
  10. 2023Learning to Compress Prompts with Gist Tokenscited by 390
  11. 2023Reasoning on Graphs: Faithful and Interpretable Large Language Model Reasoningcited by 328
  12. 2021Recursively Summarizing Books with Human Feedbackcited by 290
5 more
  1. 2024LongRAG: Enhancing Retrieval-Augmented Generation with Long-context LLMscited by 108
  2. 2023Dr.ICL: Demonstration-Retrieved In-context Learningcited by 72
  3. 2023Complex QA and language models hybrid architectures, Surveycited by 14
  4. 2024OneGen: Efficient One-Pass Unified Generation and Retrieval for LLMscited by 14
  5. 2024KnowLA: Enhancing Parameter-efficient Finetuning with Knowledgeable Adaptationcited by 10

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 12; 2026: 10.

22 papers.

Sort:

By area:

  1. 2025Synergizing RAG and Reasoning: A Systematic Reviewcited by 27
  2. 2025MemSearcher: Training LLMs to Reason, Search and Manage Memory via End-to-End Reinforcement Learningcited by 27
  3. 2026Reasoning in Trees: Improving Retrieval-Augmented Generation for Multi-Hop Question Answeringcited by 14
  4. 2025DocHop-QA: Towards Multi-Hop Reasoning over Multimodal Document Collectionscited by 8
  5. 2025When Can Large Reasoning Models Save Thinking? Mechanistic Analysis of Behavioral Divergence in Reasoningcited by 7
  6. 2025Curriculum Guided Reinforcement Learning for Efficient Multi Hop Retrieval Augmented Generationcited by 5
  7. 2025GlobalRAG: Enhancing Global Reasoning in Multi-hop Question Answering via Reinforcement Learningcited by 5
  8. 2025Think Before You Retrieve: Learning Test-Time Adaptive Search with Small Language Modelscited by 5
  9. 2025TreeHop: Efficient Embedding-Level Query Rewritercited by 3
  10. 2025HetaRAG: Hybrid Deep Retrieval-Augmented Generation across Heterogeneous Data Storescited by 3
  11. 2025StepChain GraphRAG: Reasoning Over Knowledge Graphs for Multi-Hop Question Answeringcited by 3
  12. 2025How does Watermarking Affect Visual Language Models in Document Understanding?cited by 2
10 more
  1. 2025MuISQA: Multi-Intent Retrieval-Augmented Generation for Scientific Question Answeringcited by 2
  2. 2026PAR2-RAG: Planned Active Retrieval and Reasoning for Multi-Hop Question Answeringcited by 2
  3. 2026IterCOMP: Reasoning-aware Adaptive Prompt Compression for Multi-hop Question Answeringcited by 1
  4. 2026Dense Process Supervision for Search Agents via Fact Utility Estimationcited by 1
  5. 2026MAB-DQA: Addressing Query Aspect Importance in Document Question Answering with Multi-Armed Banditscited by 0
  6. 2026SEARCH-R: Structured Entity-Aware Retrieval with Chain-of-Reasoning Navigator for Multi-hop Question Answeringcited by 0
  7. 2026D2F-ReAG: Dynamic Decomposition and Filtering for Multi-Hop Reasoning-Augmented Generationcited by 0
  8. 2026HyperProve: Answer-Guided Hypergraph Expansion for Multi-Hop Question Answeringcited by 0
  9. 2026Beyond Similarity through Zero-Token Geometric Graphs for Multi-Hop RAGcited by 0
  10. 2026ANTMAN: Adaptive Need Tracking for Multi-Agent Navigation in Large Information Spacescited by 0

Similar, not cited

Close in wording, no citation either way

  1. 2024Hierarchical Retrieval-Augmented Generation Model with Rethink for Multi-hop Question Answering
  2. 2024EfficientRAG: Efficient Retriever for Multi-Hop Question Answering
  3. 2024Tree of Reviews: A Tree-based Dynamic Iterative Retrieval Framework for Multi-hop Question Answering
  4. 2025Optimizing Question Semantic Space for Dynamic Retrieval-Augmented Multi-hop Question Answering
  5. 2026Don't Lose Entities from Retrieval to Generation: Dual Entity Recovery RAG for multi-hop QA
  6. 2026BridgeRAG: Training-Free Bridge-Conditioned Retrieval for Multi-Hop Question Answering
  7. 2026A Matryoshka Hierarchical RAG for Efficient Multi-Hop Question Answering
  8. 2026ConRAG: Consensus-Driven Multi-View Retrieval for Multi-Hop Question Answering