# A Hybrid Neural Network Model for Commonsense Reasoning

arXiv: https://arxiv.org/abs/1907.11983 · HTML: https://mlsanity.com/paper/1907.11983/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 Jul 2019.

## Built on by (most cited first, 10 of 10)

- [How Robust is GPT-3.5 to Predecessors? A Comprehensive Study on Language Understanding Tasks](https://mlsanity.com/paper/2303.00293/) (2023)
- [The Defeat of the Winograd Schema Challenge](https://mlsanity.com/paper/2201.02387/) (2022)
- [The Box is in the Pen: Evaluating Commonsense Reasoning in Neural Machine Translation](https://mlsanity.com/paper/2503.03308/) (2025)
- [Unsupervised Pronoun Resolution via Masked Noun-Phrase Prediction](https://mlsanity.com/paper/2105.12392/) (2021)
- [Enhancing Natural Language Representation with Large-Scale Out-of-Domain Commonsense](https://mlsanity.com/paper/2109.02572/) (2021)
- [Attention-based Contrastive Learning for Winograd Schemas](https://mlsanity.com/paper/2109.05108/) (2021)
- [Integrating Symbolic Reasoning into Neural Generative Models for Design Generation](https://mlsanity.com/paper/2310.09383/) (2023)
- [PronounFlow: A Hybrid Approach for Calibrating Pronouns in Sentences](https://mlsanity.com/paper/2308.15235/) (2023)
- [Meta-RTL: Reinforcement-Based Meta-Transfer Learning for Low-Resource Commonsense Reasoning](https://mlsanity.com/paper/2409.19075/) (2024)
- [MetaXCR: Reinforcement-Based Meta-Transfer Learning for Cross-Lingual Commonsense Reasoning](https://mlsanity.com/paper/2503.06531/) (2025)

## Abstract (arXiv)

This paper proposes a hybrid neural network (HNN) model for commonsense reasoning. An HNN consists of two component models, a masked language model and a semantic similarity model, which share a BERT-based contextual encoder but use different model-specific input and output layers. HNN obtains new state-of-the-art results on three classic commonsense reasoning tasks, pushing the WNLI benchmark to 89%, the Winograd Schema Challenge (WSC) benchmark to 75.1%, and the PDP60 benchmark to 90.0%. An ablation study shows that language models and semantic similarity models are complementary approaches to commonsense reasoning, and HNN effectively combines the strengths of both. The code and pre-trained models will be publicly available at https://github.com/namisan/mt-dnn.