mlsanityData to 14 Oct 2026

Multi-hop Reading Comprehension through Question Decomposition and Rescoring

arXiv 1906.02916 ↗First posted 7 Jun 2019cs.CL cs.AIcited by 119

Not on the map. Why?

Cited by 33 in LLM reasoning, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

Multi-hop Reading Comprehension (RC) requires reasoning and aggregation across several paragraphs. We propose a system for multi-hop RC that decomposes a compositional question into simpler sub-questions that can be answered by off-the-shelf single-hop RC models. Since annotations for such decomposition are expensive, we recast sub-question generation as a span prediction problem and show that our method, trained using only 400 labeled examples, generates sub-questions that are as effective as human-authored sub-questions. We also introduce a new global rescoring approach that considers each decomposition (i.e. the sub-questions and their answers) to select the best final answer, greatly improving overall performance. Our experiments on HotpotQA show that this approach achieves the state-of-the-art results, while providing explainable evidence for its decision making in the form of sub-questions.

Key facts. Cited by 119 papers in the map (data through 14 Oct 2026). First posted on arXiv 7 Jun 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 23; 2022: 30; 2023: 30; 2024: 15; 2025: 12; 2026: 9.

119 papers.

Sort:

By area:

  1. 2023Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Promptingcited by 1,122
  2. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,075
  3. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 916
  4. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 743
  5. 2022Towards Reasoning in Large Language Models: A Surveycited by 596
  6. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 558
  7. 2023Augmented Language Models: a Surveycited by 403
  8. 2022Demonstrate-Search-Predict: Composing retrieval and language models for knowledge-intensive NLPcited by 318
  9. 2023ADaPT: As-Needed Decomposition and Planning with Language Modelscited by 177
  10. 2022Internet-augmented language models through few-shot prompting for open-domain question answeringcited by 131
  11. 2023Complex Claim Verification with Evidence Retrieved in the Wildcited by 110
  12. 2023Question Decomposition Improves the Faithfulness of Model-Generated Reasoningcited by 108
38 more
  1. 2021Case-based Reasoning for Natural Language Queries over Knowledge Basescited by 107
  2. 2024Approaching Human-Level Forecasting with Language Modelscited by 107
  3. 2021Improving Multi-hop Knowledge Base Question Answering by Learning Intermediate Supervision Signalscited by 101
  4. 2023Knowledge-Driven CoT: Exploring Faithful Reasoning in LLMs for Knowledge-intensive Question Answeringcited by 97
  5. 2023Exploring Chain-of-Thought Style Prompting for Text-to-SQLcited by 69
  6. 2023Natural Language Reasoning, A Surveycited by 61
  7. 2022Generating Natural Language Proofs with Verifier-Guided Searchcited by 52
  8. 2022Is a Question Decomposition Unit All We Need?cited by 37
  9. 2023Explainable Claim Verification via Knowledge-Grounded Reasoning with Large Language Modelscited by 37
  10. 2024MoreHopQA: More Than Multi-hop Reasoningcited by 37
  11. 2021FaVIQ: FAct Verification from Information-seeking Questionscited by 36
  12. 2021Decomposing Complex Questions Makes Multi-Hop QA Easier and More Interpretablecited by 35
  13. 2022Natural Language Deduction through Search over Statement Compositionscited by 35
  14. 2021Guiding the Growth: Difficulty-Controllable Question Generation through Step-by-Step Rewritingcited by 30
  15. 2023The ART of LLM Refinement: Ask, Refine, and Trustcited by 29
  16. 2024Cofca: A Step-Wise Counterfactual Multi-hop QA benchmarkcited by 27
  17. 2025HopRAG: Multi-Hop Reasoning for Logic-Aware Retrieval-Augmented Generationcited by 24
  18. 2023Causal Reasoning of Entities and Events in Procedural Textscited by 23
  19. 2021Rissanen Data Analysis: Examining Dataset Characteristics via Description Lengthcited by 22
  20. 2023The Art of SOCRATIC QUESTIONING: Recursive Thinking with Large Language Modelscited by 22
  21. 2022Reasoning over Public and Private Data in Retrieval-Based Systemscited by 21
  22. 2021Database Reasoning Over Textcited by 20
  23. 2021MuMuQA: Multimedia Multi-Hop News Question Answering via Cross-Media Knowledge Extraction and Groundingcited by 19
  24. 2022Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Textscited by 18
  25. 2024Chain of Logic: Rule-Based Reasoning with Large Language Modelscited by 18
  26. 2024Divide-or-Conquer? Which Part Should You Distill Your LLM?cited by 17
  27. 2024Metasql: A Generate-then-Rank Framework for Natural Language to SQL Translationcited by 17
  28. 2024LLMs Know What They Need: Leveraging a Missing Information Guided Framework to Empower Retrieval-Augmented Generationcited by 17
  29. 2023SCREWS: A Modular Framework for Reasoning with Revisionscited by 16
  30. 2022Measuring Compositional Consistency for Video Question Answeringcited by 15
  31. 2023Question Decomposition Tree for Answering Complex Questions over Knowledge Basescited by 15
  32. 2025MemoTime: Memory-Augmented Temporal Knowledge Graph Enhanced Large Language Model Reasoningcited by 15
  33. 2021Complex Factoid Question Answering with a Free-Text Knowledge Graphcited by 14
  34. 2023Semantic Decomposition of Question and SQL for Text-to-SQL Parsingcited by 14
  35. 2021Graph-free Multi-hop Reading Comprehension: A Select-to-Guide Strategycited by 13
  36. 2021Hey AI, Can You Solve Complex Tasks by Talking to Agents?cited by 13
  37. 2022From Easy to Hard: Two-stage Selector and Reader for Multi-hop Question Answeringcited by 12
  38. 2022Interpretable AMR-Based Question Decomposition for Multi-hop Question Answeringcited by 12

50 most cited of 119 All 119 (JSON)