mlsanityData to 7 Oct 2026

Text Modular Networks: Learning to Decompose Tasks in the Language of Existing Models

arXiv 2009.00751 ↗First posted 1 Sep 2020cs.CL cs.AIcited by 65

Not on the map. Why?

Cited by 20 in LLM reasoning, but placed by its own abstract, which didn’t fit.

Abstract (arXiv)

We propose a general framework called Text Modular Networks(TMNs) for building interpretable systems that learn to solve complex tasks by decomposing them into simpler ones solvable by existing models. To ensure solvability of simpler tasks, TMNs learn the textual input-output behavior (i.e., language) of existing models through their datasets. This differs from prior decomposition-based approaches which, besides being designed specifically for each complex task, produce decompositions independent of existing sub-models. Specifically, we focus on Question Answering (QA) and show how to train a next-question generator to sequentially produce sub-questions targeting appropriate sub-models, without additional human annotation. These sub-questions and answers provide a faithful natural language explanation of the model's reasoning. We use this framework to build ModularQA, a system that can answer multi-hop reasoning questions by decomposing them into sub-questions answerable by a neural factoid single-span QA model and a symbolic calculator. Our experiments show that ModularQA is more versatile than existing explainable systems for DROP and HotpotQA datasets, is more robust than state-of-the-art blackbox (uninterpretable) systems, and generates more understandable and trustworthy explanations compared to prior work.

Key facts. Cited by 65 papers in the map (data through 7 Oct 2026). First posted on arXiv 1 Sep 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 13; 2022: 19; 2023: 20; 2024: 4; 2025: 7; 2026: 2.

65 papers.

Sort:

By area:

  1. 2022Measuring and Narrowing the Compositionality Gap in Language Modelscited by 1,054
  2. 2021MuSiQue: Multihop Questions via Single-hop Question Compositioncited by 893
  3. 2021Did Aristotle Use a Laptop? A Question Answering Benchmark with Implicit Reasoning Strategiescited by 736
  4. 2022Decomposed Prompting: A Modular Approach for Solving Complex Taskscited by 554
  5. 2023Chameleon: Plug-and-Play Compositional Reasoning with Large Language Modelscited by 352
  6. 2022Binding Language Models in Symbolic Languagescited by 224
  7. 2023ADaPT: As-Needed Decomposition and Planning with Language Modelscited by 172
  8. 2021CrossFit: A Few-shot Learning Challenge for Cross-task Generalization in NLPcited by 142
  9. 2023Branch-Solve-Merge Improves Large Language Model Evaluation and Generationcited by 104
  10. 2023Knowledge-Driven CoT: Exploring Faithful Reasoning in LLMs for Knowledge-intensive Question Answeringcited by 97
  11. 2022Successive Prompting for Decomposing Complex Questionscited by 94
  12. 2023Neurons in Large Language Models: Dead, N-gram, Positionalcited by 88
38 more
  1. 2021Prompt Waywardness: The Curious Case of Discretized Interpretation of Continuous Promptscited by 64
  2. 2023Natural Language Reasoning, A Surveycited by 61
  3. 2022Reasoning Like Program Executorscited by 45
  4. 2021Turning Tables: Generating Examples from Semi-structured Tables for Endowing Language Models with Reasoning Skillscited by 39
  5. 2023Forward-Backward Reasoning in Large Language Models for Mathematical Verificationcited by 39
  6. 2022Is a Question Decomposition Unit All We Need?cited by 37
  7. 2021Reason first, then respond: Modular Generation for Knowledge-infused Dialoguecited by 32
  8. 2021Guiding the Growth: Difficulty-Controllable Question Generation through Step-by-Step Rewritingcited by 30
  9. 2021Flexible Generation of Natural Language Deductionscited by 28
  10. 2023Automating the Generation of Prompts for LLM-based Action Choice in PDDL Planningcited by 24
  11. 2023Instruction Tuned Models are Quick Learnerscited by 21
  12. 2021Factoring Statutory Reasoning as Language Understanding Challengescited by 20
  13. 2021Pushing the Limits of Rule Reasoning in Transformers through Natural Language Satisfiabilitycited by 18
  14. 2022Learning to Decompose: Hypothetical Question Decomposition Based on Comparable Textscited by 18
  15. 2024Chain of Logic: Rule-Based Reasoning with Large Language Modelscited by 18
  16. 2022HELP ME THINK: A Simple Prompting Strategy for Non-experts to Create Customized Content with Modelscited by 17
  17. 2023GEAR: Augmenting Language Models with Generalizable and Efficient Tool Resolutioncited by 17
  18. 2024LLMs Know What They Need: Leveraging a Missing Information Guided Framework to Empower Retrieval-Augmented Generationcited by 17
  19. 2022NELLIE: A Neuro-Symbolic Inference Engine for Grounded, Compositional, and Explainable Reasoningcited by 16
  20. 2021Teaching Autoregressive Language Models Complex Tasks By Demonstrationcited by 15
  21. 2024Enhancing LLM Reasoning with Multi-Path Collaborative Reactive and Reflection agentscited by 15
  22. 2022Summarization Programs: Interpretable Abstractive Summarization with Neural Modular Treescited by 14
  23. 2021Hey AI, Can You Solve Complex Tasks by Talking to Agents?cited by 13
  24. 2021Truth-Conditional Captioning of Time Series Datacited by 11
  25. 2023DORIS-MAE: Scientific Document Retrieval using Multi-level Aspect-based Queriescited by 11
  26. 2023Exploring Question Decomposition for Zero-Shot VQAcited by 11
  27. 2023Natural Language Decomposition and Interpretation of Complex Utterancescited by 10
  28. 2022Calibrating Trust of Multi-Hop Question Answering Systems with Decompositional Probescited by 9
  29. 2022Inferring Implicit Relations in Complex Questions with Language Modelscited by 9
  30. 2022Teaching Broad Reasoning Skills for Multi-Step QA by Generating Hard Contextscited by 7
  31. 2022MURMUR: Modular Multi-Step Reasoning for Semi-Structured Data-to-Text Generationcited by 7
  32. 2025Position: Scaling LLM Agents Requires Asymptotic Analysis with LLM Primitivescited by 7
  33. 2025Fast-Slow-Thinking: Complex Task Solving with Large Language Modelscited by 7
  34. 2022Locate Then Ask: Interpretable Stepwise Reasoning for Multi-hop Question Answeringcited by 6
  35. 2022Breakpoint Transformers for Modeling and Tracking Intermediate Beliefscited by 6
  36. 2023HPE:Answering Complex Questions over Text by Hybrid Question Parsing and Executioncited by 5
  37. 2025POQD: Performance-Oriented Query Decomposer for Multi-vector retrievalcited by 5
  38. 2024DEXTER: A Benchmark for open-domain Complex Question Answering using LLMscited by 4

50 most cited of 65 All 65 (JSON)