mlsanityData to 7 Oct 2026

Improving Numerical Reasoning Skills in the Modular Approach for Complex Question Answering on Text

arXiv 2109.02289 ↗First posted 6 Sep 2021cs.CL cs.AIcited by 2

Not on the map. Why?

Abstract (arXiv)

Numerical reasoning skills are essential for complex question answering (CQA) over text. It requires opertaions including counting, comparison, addition and subtraction. A successful approach to CQA on text, Neural Module Networks (NMNs), follows the programmer-interpreter paradigm and leverages specialised modules to perform compositional reasoning. However, the NMNs framework does not consider the relationship between numbers and entities in both questions and paragraphs. We propose effective techniques to improve NMNs' numerical reasoning capabilities by making the interpreter question-aware and capturing the relationship between entities and numbers. On the same subset of the DROP dataset for CQA on text, experimental results show that our additions outperform the original NMNs by 3.0 points for the overall F1 score.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 12 of its 13 references that are in the map. First posted on arXiv 6 Sep 2021.

Builds on

12 of 13 references.

References · 12

  1. 2019DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphscited by 883
  2. 2015Neural Module Networkscited by 375
  3. 2016Bidirectional Attention Flow for Machine Comprehensioncited by 278
  4. 2018QANet: Combining Local Convolution with Global Self-Attention for Reading Comprehensioncited by 164
  5. 2019Neural Module Networks for Reasoning over Textcited by 65
  6. 2019NumNet: Machine Reading Comprehension with Numerical Reasoningcited by 59
  7. 2019A Multi-Type Multi-Span Network for Reading Comprehension that Requires Discrete Reasoningcited by 41
  8. 2020Question Directed Graph Attention Network for Numerical Reasoning over Textcited by 35
  9. 2020Few-Shot Complex Knowledge Base Question Answering via Meta Reinforcement Learningcited by 16
  10. 2020Modeling Topical Relevance for Multi-Turn Dialogue Generationcited by 15
  11. 2021Weakly Supervised Neuro-Symbolic Module Networks for Numerical Reasoningcited by 10
  12. 2020Understanding Unnatural Questions Improves Reasoning over Textcited by 1

Built on by

Cited by year

Citing papers per year: 2022: 2.

2 papers.

  1. 2022NAPG: Non-Autoregressive Program Generation for Hybrid Tabular-Textual Question Answeringcited by 4
  2. 2022Teaching Neural Module Networks to Do Arithmeticcited by 1