# Unsupervised Natural Language Inference via Decoupled Multimodal Contrastive Learning

arXiv: https://arxiv.org/abs/2010.08200 · HTML: https://mlsanity.com/paper/2010.08200/

## Key facts

- Cited by 8 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 16 Oct 2020.

## Built on by (most cited first, 8 of 8)

- [MultiBench: Multiscale Benchmarks for Multimodal Representation Learning](https://mlsanity.com/paper/2107.07502/) (2021)
- [CLEVE: Contrastive Pre-training for Event Extraction](https://mlsanity.com/paper/2105.14485/) (2021)
- [MLIP: Enhancing Medical Visual Representation with Divergence Encoder and Knowledge-guided Contrastive Learning](https://mlsanity.com/paper/2402.02045/) (2024)
- [Multilingual Molecular Representation Learning via Contrastive Pre-training](https://mlsanity.com/paper/2109.08830/) (2021)
- [SANCL: Multimodal Review Helpfulness Prediction with Selective Attention and Natural Contrastive Learning](https://mlsanity.com/paper/2209.05040/) (2022)
- [Unsupervised Natural Language Inference Using PHL Triplet Generation](https://mlsanity.com/paper/2110.08438/) (2021)
- [Contrastive Cross-Modal Knowledge Sharing Pre-training for Vision-Language Representation Learning and Retrieval](https://mlsanity.com/paper/2207.00733/) (2022)
- [FSMR: A Feature Swapping Multi-modal Reasoning Approach with Joint Textual and Visual Clues](https://mlsanity.com/paper/2403.20026/) (2024)

## Abstract (arXiv)

We propose to solve the natural language inference problem without any supervision from the inference labels via task-agnostic multimodal pretraining. Although recent studies of multimodal self-supervised learning also represent the linguistic and visual context, their encoders for different modalities are coupled. Thus they cannot incorporate visual information when encoding plain text alone. In this paper, we propose Multimodal Aligned Contrastive Decoupled learning (MACD) network. MACD forces the decoupled text encoder to represent the visual information via contrastive learning. Therefore, it embeds visual knowledge even for plain text inference. We conducted comprehensive experiments over plain text inference datasets (i.e. SNLI and STS-B). The unsupervised MACD even outperforms the fully-supervised BiLSTM and BiLSTM+ELMO on STS-B.