mlsanityData to 7 Oct 2026

Chain-of-Questions Training with Latent Answers for Robust Multistep Question Answering

arXiv 2305.14901 ↗First posted 24 May 2023cs.CL cs.LGcited by 8

Not on the map. Why?

Abstract (arXiv)

We train a language model (LM) to robustly answer multistep questions by generating and answering sub-questions. We propose Chain-of-Questions, a framework that trains a model to generate sub-questions and sub-answers one at a time by leveraging human annotated question decomposition meaning representation (QDMR). The key technical challenge is that QDMR only contains sub-questions but not answers to those sub-questions, so we treat sub-answers as latent variables and optimize them using a novel dynamic mixture of Hard-EM and MAPO. Chain-of-Questions greatly outperforms strong neuro-symbolic methods by 9.0 F1 on DROP contrast set, and outperforms GPT-3.5 by 24.3 F1 on HOTPOTQA adversarial set, thus demonstrating the effectiveness and robustness of our framework.

Key facts. Cited by 8 papers in the map (data through 7 Oct 2026). Builds on 30 of its 37 references that are in the map. First posted on arXiv 24 May 2023.

Builds on

30 of 37 references.

References · 30

  1. 2020Language Models are Few-Shot Learnerscited by 32,199
  2. 2019Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformercited by 14,491
  3. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,302
  4. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  5. 2020Longformer: The Long-Document Transformercited by 3,362
  6. 2022STaR: Bootstrapping Reasoning With Reasoningcited by 895
  7. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 893
  8. 2019DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphscited by 883
  9. 2015Scheduled Sampling for Sequence Prediction with Recurrent Neural Networkscited by 779
  10. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
  11. 2021FinQA: A Dataset of Numerical Reasoning over Financial Datacited by 552
  12. 2020UnifiedQA: Crossing Format Boundaries With a Single QA Systemcited by 475
18 more
  1. 2018The Web as a Knowledge-base for Answering Complex Questionscited by 409
  2. 2021LongT5: Efficient Text-To-Text Transformer for Long Sequencescited by 232
  3. 2019MRQA 2019 Shared Task: Evaluating Generalization in Reading Comprehensioncited by 183
  4. 2020Break It Down: A Question Understanding Benchmarkcited by 131
  5. 2019Multi-hop Reading Comprehension through Question Decomposition and Rescoringcited by 119
  6. 2020Unsupervised Question Decomposition for Question Answeringcited by 118
  7. 2022Successive Prompting for Decomposing Complex Questionscited by 94
  8. 2019Reasoning Over Paragraph Effects in Situationscited by 79
  9. 2019A Discrete Hard EM Approach for Weakly Supervised Question Answeringcited by 78
  10. 2019Avoiding Reasoning Shortcuts: Adversarial Evaluation, Training, and Model Development for Multi-Hop QAcited by 63
  11. 2018Memory Augmented Policy Optimization for Program Synthesis and Semantic Parsingcited by 57
  12. 2019Giving BERT a Calculator: Finding Operations and Arguments with Reading Comprehensioncited by 51
  13. 2019Learning to Generalize from Sparse and Underspecified Rewardscited by 39
  14. 2021Decomposing Complex Questions Makes Multi-Hop QA Easier and More Interpretablecited by 35
  15. 2020A Simple Yet Strong Pipeline for HotpotQAcited by 24
  16. 2020Semantics Altering Modifications for Evaluating Comprehension in Machine Readingcited by 13
  17. 2022Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contextscited by 7
  18. 2022Complex Reading Comprehension Through Question Decompositioncited by 7

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 3; 2025: 3.

8 papers.

  1. 2024Divide-or-Conquer? Which Part Should You Distill Your LLM?cited by 17
  2. 2023Self-Contradictory Reasoning Evaluation and Detectioncited by 14
  3. 2025Route-and-Reason: Scaling Large Language Model Reasoning with Reinforced Model Routercited by 13
  4. 2024Language Models can Infer Action Semantics for Symbolic Planners from Environment Feedbackcited by 8
  5. 2023Efficient End-to-End Visual Document Understanding with Rationale Distillationcited by 5
  6. 2025POQD: Performance-Oriented Query Decomposer for Multi-vector retrievalcited by 5
  7. 2025Cooperative Retrieval-Augmented Generation for Question Answering: Mutual Information Exchange and Ranking by Contrasting Layerscited by 2
  8. 2024A Survey of Query Optimization in Large Language Modelscited by 1