• Corpus ID: 209202200

Neural Module Networks for Reasoning over Text

  title={Neural Module Networks for Reasoning over Text},
  author={Nitish Gupta and Kevin Lin and Dan Roth and Sameer Singh and Matt Gardner},
Answering compositional questions that require multiple steps of reasoning against text is challenging, especially when they involve discrete, symbolic operations. Neural module networks (NMNs) learn to parse such questions as executable programs composed of learnable modules, performing well on synthetic visual QA domains. However, we find that it is challenging to learn these models for non-synthetic questions on open-domain text, where a model needs to deal with the diversity of natural… 

Figures and Tables from this paper

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

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.

How to Reuse and Compose Knowledge for a Lifetime of Tasks: A Survey on Continual Learning and Functional Composition

A major goal of artificial intelligence (AI) is to create an agent capable of acquiring a general understanding of the world. Such an agent would require the ability to continually accumulate and

Weakly Supervised Neuro-Symbolic Module Networks for Numerical Reasoning

Weakly Supervised Neuro-Symbolic Module Network (WNSMN) trained with answers as the sole supervision for numerical reasoning based MRC outperforms NMN by 32% and the reasoning-free generative language model GenBERT by 8% in exact match accuracy under comparable weakly supervised settings.

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

This work proposes effective techniques to improve NMNs’ numerical reasoning capabilities by making the interpreter questionaware and capturing the relationship between entities and numbers.

Toward Code Generation: A Survey and Lessons from Semantic Parsing

An overview of the growing body of research in semantic parsing works from an evolutionary perspective, with specific analyses on neuro-symbolic methods, architecture, and supervision is presented.

Understanding Unnatural Questions Improves Reasoning over Text

This paper addresses the challenge of learning a high-quality programmer (parser) by projecting natural human-generated questions into unnatural machine-generated Questions which are more convenient to parse by learning a semantic parser that associates synthetic questions with their corresponding action sequences.

Obtaining Faithful Interpretations from Compositional Neural Networks

It is found that the intermediate outputs of NMNs differ from the expected output, illustrating that the network structure does not provide a faithful explanation of model behaviour, and particular choices for module architecture are proposed that yield much better faithfulness, at a minimal cost to accuracy.

On the Explainability of Natural Language Processing Deep Models

A survey that studies model-agnostic as well as model-specific explainability methods on NLP models, and decomposes the methods into three categories according to what they explain, to democratize ExAI methods in the NLP field.

Deep Inductive Logic Reasoning for Multi-Hop Reading Comprehension

A deep-learning based inductive logic reasoning method that firstly extracts query-related (candidate-related) information, and then conducts logic reasoning among the filtered information by inducing feasible rules that entail the target relation is proposed.

Learning to Reason Deductively: Math Word Problem Solving as Complex Relation Extraction

This work views the task as a complex relation extraction problem, and proposes a novel approach that presents explainable deductive reasoning steps to iteratively construct target expressions, where each step involves a primitive operation over two quantities defining their relation.



A Multi-Type Multi-Span Network for Reading Comprehension that Requires Discrete Reasoning

The Multi-Type Multi-Span Network (MTMSN) is introduced, a neural reading comprehension model that combines a multi-type answer predictor designed to support various answer types with amulti-span extraction method for dynamically producing one or multiple text spans.

DROP: A Reading Comprehension Benchmark Requiring Discrete Reasoning Over Paragraphs

A new reading comprehension benchmark, DROP, which requires Discrete Reasoning Over the content of Paragraphs, and presents a new model that combines reading comprehension methods with simple numerical reasoning to achieve 51% F1.

CLEVR: A Diagnostic Dataset for Compositional Language and Elementary Visual Reasoning

This work presents a diagnostic dataset that tests a range of visual reasoning abilities and uses this dataset to analyze a variety of modern visual reasoning systems, providing novel insights into their abilities and limitations.

Self-Assembling Modular Networks for Interpretable Multi-Hop Reasoning

This work presents an interpretable, controller-based Self-Assembling Neural Modular Network for multi-hop reasoning, where four novel modules (Find, Relocate, Compare, NoOp) are designed to perform unique types of language reasoning.

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

This work enables a BERT-based reading comprehension model to perform lightweight numerical reasoning by augmenting the model with a predefined set of executable ‘programs’ which encompass simple arithmetic as well as extraction.

SG-Net: Syntax-Guided Machine Reading Comprehension

This work uses syntax to guide the text modeling by incorporating explicit syntactic constraints into attention mechanism for better linguistically motivated word representations and shows that the proposed SG-Net design helps achieve substantial performance improvement over strong baselines.

The Neuro-Symbolic Concept Learner: Interpreting Scenes Words and Sentences from Natural Supervision

We propose the Neuro-Symbolic Concept Learner (NS-CL), a model that learns visual concepts, words, and semantic parsing of sentences without explicit supervision on any of them; instead, our model

CommonsenseQA: A Question Answering Challenge Targeting Commonsense Knowledge

This work presents CommonsenseQA: a challenging new dataset for commonsense question answering, which extracts from ConceptNet multiple target concepts that have the same semantic relation to a single source concept.

BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding

A new language representation model, BERT, designed to pre-train deep bidirectional representations from unlabeled text by jointly conditioning on both left and right context in all layers, which can be fine-tuned with just one additional output layer to create state-of-the-art models for a wide range of tasks.

HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answering

It is shown that HotpotQA is challenging for the latest QA systems, and the supporting facts enable models to improve performance and make explainable predictions.