mlsanityData to 14 Oct 2026

Giving BERT a Calculator: Finding Operations and Arguments with Reading Comprehension

arXiv 1909.00109 ↗First posted 31 Aug 2019cs.CL cs.LGcited by 51

Not on the map. Why?

Abstract (arXiv)

Reading comprehension models have been successfully applied to extractive text answers, but it is unclear how best to generalize these models to abstractive numerical answers. We enable a BERT-based reading comprehension model to perform lightweight numerical reasoning. We augment the model with a predefined set of executable 'programs' which encompass simple arithmetic as well as extraction. Rather than having to learn to manipulate numbers directly, the model can pick a program and execute it. On the recent Discrete Reasoning Over Passages (DROP) dataset, designed to challenge reading comprehension models, we show a 33% absolute improvement by adding shallow programs. The model can learn to predict new operations when appropriate in a math word problem setting (Roy and Roth, 2015) with very few training examples.

Key facts. Cited by 51 papers in the map (data through 14 Oct 2026). First posted on arXiv 31 Aug 2019.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2021: 12; 2022: 22; 2023: 13; 2024: 2; 2025: 1; 2026: 1.

51 papers.

Sort:

By area:

  1. 2022Chain-of-Thought Prompting Elicits Reasoning in Large Language Modelscited by 12,459
  2. 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,708
  3. 2022Least-to-Most Prompting Enables Complex Reasoning in Large Language Modelscited by 1,407
  4. 2023Gorilla: Large Language Model Connected with Massive APIscited by 1,332
  5. 2023CRITIC: Large Language Models Can Self-Correct with Tool-Interactive Critiquingcited by 783
  6. 2022PAL: Program-aided Language Modelscited by 652
  7. 2021TAT-QA: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Financecited by 408
  8. 2023Augmented Language Models: a Surveycited by 403
  9. 2022Making Large Language Models Better Reasoners with Step-Aware Verifiercited by 233
  10. 2023REFINER: Reasoning Feedback on Intermediate Representationscited by 205
  11. 2022TALM: Tool Augmented Language Modelscited by 197
  12. 2023ChatDB: Augmenting LLMs with Databases as Their Symbolic Memorycited by 140
38 more
  1. 2021Investigating the Limitations of Transformers with Simple Arithmetic Taskscited by 139
  2. 2022Prompting Is Programming: A Query Language for Large Language Modelscited by 110
  3. 2022Successive Prompting for Decomposing Complex Questionscited by 94
  4. 2021Representing Numbers in NLP: a Survey and a Visioncited by 83
  5. 2022Towards Complex Document Understanding By Discrete Reasoningcited by 82
  6. 2021MATE: Multi-view Attention for Table Transformer Efficiencycited by 54
  7. 2022Reasoning Like Program Executorscited by 45
  8. 2022PACIFIC: Towards Proactive Conversational Question Answering over Tabular and Textual Data in Financecited by 45
  9. 2021MWP-BERT: Numeracy-Augmented Pre-training for Math Word Problem Solvingcited by 32
  10. 2022Exploring The Landscape of Distributional Robustness for Question Answering Modelscited by 23
  11. 2021Database Reasoning Over Textcited by 20
  12. 2021FORTAP: Using Formulas for Numerical-Reasoning-Aware Table Pretrainingcited by 19
  13. 2024Path-Consistency with Prefix Enhancement for Efficient Inference in LLMscited by 17
  14. 2022UniRPG: Unified Discrete Reasoning over Table and Text as Program Generationcited by 13
  15. 2023ProTIP: Progressive Tool Retrieval Improves Planningcited by 12
  16. 2021NT5?! Training T5 to Perform Numerical Reasoningcited by 11
  17. 2021Truth-Conditional Captioning of Time Series Datacited by 11
  18. 2022Induced Natural Language Rationales and Interleaved Markup Tokens Enable Extrapolation in Large Language Modelscited by 10
  19. 2022A Neural-Symbolic Approach to Natural Language Understandingcited by 8
  20. 2023Chain-of-Questions Training with Latent Answers for Robust Multistep Question Answeringcited by 8
  21. 2022OPERA:Operation-Pivoted Discrete Reasoning over Textcited by 7
  22. 2022Reflection of Thought: Inversely Eliciting Numerical Reasoning in Language Models via Solving Linear Systemscited by 7
  23. 2022Generalizing Math Word Problem Solvers via Solution Diversificationcited by 7
  24. 2023A Comprehensive Evaluation of Tool-Assisted Generation Strategiescited by 6
  25. 2023Doc2SoarGraph: Discrete Reasoning over Visually-Rich Table-Text Documents via Semantic-Oriented Hierarchical Graphscited by 5
  26. 2024Procedural Adherence and Interpretability Through Neuro-Symbolic Generative Agentscited by 5
  27. 2022Arithmetic-Based Pretraining -- Improving Numeracy of Pretrained Language Modelscited by 3
  28. 2022A Survey on Knowledge-Enhanced Pre-trained Language Modelscited by 3
  29. 2023FERMAT: An Alternative to Accuracy for Numerical Reasoningcited by 3
  30. 2023Grounded Intuition of GPT-Vision's Abilities with Scientific Imagescited by 3
  31. 2026Inclusion-of-Thoughts: Mitigating Preference Instability via Purifying the Decision Spacecited by 3
  32. 2021A Reinforcement Learning Environment for Mathematical Reasoning via Program Synthesiscited by 2
  33. 2021EviDR: Evidence-Emphasized Discrete Reasoning for Reasoning Machine Reading Comprehensioncited by 2
  34. 2022Discourse-Aware Graph Networks for Textual Logical Reasoningcited by 2
  35. 2021Discrete Reasoning Templates for Natural Language Understandingcited by 1
  36. 2022Improving Downstream Task Performance by Treating Numbers as Entitiescited by 1
  37. 2023Towards Graph-hop Retrieval and Reasoning in Complex Question Answering over Textual Databasecited by 1
  38. 2023Around the GLOBE: Numerical Aggregation Question-Answering on Heterogeneous Genealogical Knowledge Graphs with Deep Neural Networkscited by 0

50 most cited of 51 All 51 (JSON)