mlsanityData to 7 Oct 2026

Unsupervised Question Decomposition for Question Answering

arXiv 2002.09758 ↗First posted 22 Feb 2020cs.CL cs.AI cs.LGcited by 118

Not on the map. Why?

Cited by 29 in LLM reasoning, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We aim to improve question answering (QA) by decomposing hard questions into simpler sub-questions that existing QA systems are capable of answering. Since labeling questions with decompositions is cumbersome, we take an unsupervised approach to produce sub-questions, also enabling us to leverage millions of questions from the internet. Specifically, we propose an algorithm for One-to-N Unsupervised Sequence transduction (ONUS) that learns to map one hard, multi-hop question to many simpler, single-hop sub-questions. We answer sub-questions with an off-the-shelf QA model and give the resulting answers to a recomposition model that combines them into a final answer. We show large QA improvements on HotpotQA over a strong baseline on the original, out-of-domain, and multi-hop dev sets. ONUS automatically learns to decompose different kinds of questions, while matching the utility of supervised and heuristic decomposition methods for QA and exceeding those methods in fluency. Qualitatively, we find that using sub-questions is promising for shedding light on why a QA system makes a prediction.

Key facts. Cited by 118 papers in the map (data through 7 Oct 2026). First posted on arXiv 22 Feb 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 23; 2022: 28; 2023: 26; 2024: 18; 2025: 15; 2026: 8.

118 papers.

Sort:

By area:

  1. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,397
  2. 2023Language Models Don't Always Say What They Think: Unfaithful Explanations in Chain-of-Thought Promptingcited by 1,081
  3. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,054
  4. 2022Solving math word problems with process- and outcome-based feedbackcited by 871
  5. 2022Discovering Language Model Behaviors with Model-Written Evaluationscited by 805
  6. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 736
  7. 2022Towards Reasoning in Large Language Models: A Surveycited by 593
  8. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
  9. 2023Augmented Language Models: a Surveycited by 398
  10. 2021True Few-Shot Learning with Language Modelscited by 343
  11. 2021Recursively Summarizing Books with Human Feedbackcited by 287
  12. 2024Chain of Agents: Large Language Models Collaborating on Long-Context Taskscited by 201
38 more
  1. 2023Large Language Models are Versatile Decomposers: Decompose Evidence and Questions for Table-based Reasoningcited by 177
  2. 2023ADaPT: As-Needed Decomposition and Planning with Language Modelscited by 172
  3. 2024Distilling System 2 into System 1cited by 152
  4. 2022Internet-augmented language models through few-shot prompting for open-domain question answeringcited by 131
  5. 2024SceneCraft: An LLM Agent for Synthesizing 3D Scene as Blender Codecited by 125
  6. 2022MultiHiertt: Numerical Reasoning over Multi Hierarchical Tabular and Textual Datacited by 120
  7. 2023Complex Claim Verification with Evidence Retrieved in the Wildcited by 110
  8. 2021Case-based Reasoning for Natural Language Queries over Knowledge Basescited by 107
  9. 2023Question Decomposition Improves the Faithfulness of Model-Generated Reasoningcited by 107
  10. 2024Easy-to-Hard Generalization: Scalable Alignment Beyond Human Supervisioncited by 95
  11. 2025OctoTools: An Agentic Framework with Extensible Tools for Complex Reasoningcited by 69
  12. 2023Natural Language Reasoning, A Surveycited by 61
  13. 2024Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More?cited by 53
  14. 2024RA-ISF: Learning to Answer and Understand from Retrieval Augmentation via Iterative Self-Feedbackcited by 46
  15. 2022Automatic Generation of Socratic Subquestions for Teaching Math Word Problemscited by 42
  16. 2021Evidentiality-guided Generation for Knowledge-Intensive NLP Taskscited by 40
  17. 2022Is a Question Decomposition Unit All We Need?cited by 37
  18. 2021Decomposing Complex Questions Makes Multi-Hop QA Easier and More Interpretablecited by 35
  19. 2021Select, Substitute, Search: A New Benchmark for Knowledge-Augmented Visual Question Answeringcited by 33
  20. 2025TreeReview: A Dynamic Tree of Questions Framework for Deep and Efficient LLM-based Scientific Peer Reviewcited by 31
  21. 2022Generative Language Models for Paragraph-Level Question Generationcited by 28
  22. 2024HOLMES: Hyper-Relational Knowledge Graphs for Multi-hop Question Answering using LLMscited by 24
  23. 2022Visconde: Multi-document QA with GPT-3 and Neural Rerankingcited by 23
  24. 2021Rissanen Data Analysis: Examining Dataset Characteristics via Description Lengthcited by 22
  25. 2022Reasoning over Public and Private Data in Retrieval-Based Systemscited by 21
  26. 2022Few-shot Reranking for Multi-hop QA via Language Model Promptingcited by 19
  27. 2022Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Textscited by 18
  28. 2024Divide-or-Conquer? Which Part Should You Distill Your LLM?cited by 17
  29. 2022Measuring Compositional Consistency for Video Question Answeringcited by 15
  30. 2023An Empirical Comparison of LM-based Question and Answer Generation Methodscited by 15
  31. 2025Fast on the Easy, Deep on the Hard: Efficient Reasoning via Powered Length Penaltycited by 15
  32. 2025Speak Easy: Eliciting Harmful Jailbreaks from LLMs with Simple Interactionscited by 14
  33. 2021Graph-free Multi-hop Reading Comprehension: A Select-to-Guide Strategycited by 13
  34. 2021Hey AI, Can You Solve Complex Tasks by Talking to Agents?cited by 13
  35. 2023Semantic Decomposition of Question and SQL for Text-to-SQL Parsingcited by 13
  36. 2022From Easy to Hard: Two-stage Selector and Reader for Multi-hop Question Answeringcited by 12
  37. 2022Interpretable AMR-Based Question Decomposition for Multi-hop Question Answeringcited by 12
  38. 2021Exploring Decomposition for Table-based Fact Verificationcited by 11

50 most cited of 118 All 118 (JSON)