# Unsupervised Natural Language Inference Using PHL Triplet Generation

arXiv: https://arxiv.org/abs/2110.08438 · HTML: https://mlsanity.com/paper/2110.08438/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- Builds on 34 of its 48 references that are in the map.
- First posted on arXiv 16 Oct 2021.

## Builds on (most cited first, 30 of 34)

- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [A Baseline for Detecting Misclassified and Out-of-Distribution Examples in Neural Networks](https://mlsanity.com/paper/1610.02136/) (2016)
- [ViLBERT: Pretraining Task-Agnostic Visiolinguistic Representations for Vision-and-Language Tasks](https://mlsanity.com/paper/1908.02265/) (2019)
- [A large annotated corpus for learning natural language inference](https://mlsanity.com/paper/1508.05326/) (2015)
- [SuperGLUE: A Stickier Benchmark for General-Purpose Language Understanding Systems](https://mlsanity.com/paper/1905.00537/) (2019)
- [LXMERT: Learning Cross-Modality Encoder Representations from Transformers](https://mlsanity.com/paper/1908.07490/) (2019)
- [Flickr30k Entities: Collecting Region-to-Phrase Correspondences for Richer Image-to-Sentence Models](https://mlsanity.com/paper/1505.04870/) (2015)
- [PEGASUS: Pre-training with Extracted Gap-sentences for Abstractive Summarization](https://mlsanity.com/paper/1912.08777/) (2019)
- [ConceptNet 5.5: An Open Multilingual Graph of General Knowledge](https://mlsanity.com/paper/1612.03975/) (2016)
- [Adversarial NLI: A New Benchmark for Natural Language Understanding](https://mlsanity.com/paper/1910.14599/) (2019)
- [Improving Neural Machine Translation Models with Monolingual Data](https://mlsanity.com/paper/1511.06709/) (2015)
- [Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learning](https://mlsanity.com/paper/1908.02983/) (2019)
- [Dynabench: Rethinking Benchmarking in NLP](https://mlsanity.com/paper/2104.14337/) (2021)
- [A Corpus and Evaluation Framework for Deeper Understanding of Commonsense Stories](https://mlsanity.com/paper/1604.01696/) (2016)
- [Dialogue Natural Language Inference](https://mlsanity.com/paper/1811.00671/) (2018)
- [Breaking NLI Systems with Sentences that Require Simple Lexical Inferences](https://mlsanity.com/paper/1805.02266/) (2018)
- [Training Question Answering Models From Synthetic Data](https://mlsanity.com/paper/2002.09599/) (2020)
- [Towards Zero-Label Language Learning](https://mlsanity.com/paper/2109.09193/) (2021)
- [Adversarial VQA: A New Benchmark for Evaluating the Robustness of VQA Models](https://mlsanity.com/paper/2106.00245/) (2021)
- [Unsupervised Question Answering by Cloze Translation](https://mlsanity.com/paper/1906.04980/) (2019)
- [Human-Adversarial Visual Question Answering](https://mlsanity.com/paper/2106.02280/) (2021)
- [VQA-LOL: Visual Question Answering under the Lens of Logic](https://mlsanity.com/paper/2002.08325/) (2020)
- [Template-Based Question Generation from Retrieved Sentences for Improved Unsupervised Question Answering](https://mlsanity.com/paper/2004.11892/) (2020)
- [Self-supervised Knowledge Triplet Learning for Zero-shot Question Answering](https://mlsanity.com/paper/2005.00316/) (2020)
- [WeaQA: Weak Supervision via Captions for Visual Question Answering](https://mlsanity.com/paper/2012.02356/) (2020)
- [Simple and Effective Semi-Supervised Question Answering](https://mlsanity.com/paper/1804.00720/) (2018)
- [Self-Supervised Test-Time Learning for Reading Comprehension](https://mlsanity.com/paper/2103.11263/) (2021)

## Built on by (most cited first, 6 of 6)

- [Model Cascading: Towards Jointly Improving Efficiency and Accuracy of NLP Systems](https://mlsanity.com/paper/2210.05528/) (2022)
- [Logical Satisfiability of Counterfactuals for Faithful Explanations in NLI](https://mlsanity.com/paper/2205.12469/) (2022)
- [Can Open-Domain QA Reader Utilize External Knowledge Efficiently like Humans?](https://mlsanity.com/paper/2211.12707/) (2022)
- [PECO: Examining Single Sentence Label Leakage in Natural Language Inference Datasets through Progressive Evaluation of Cluster Outliers](https://mlsanity.com/paper/2112.09237/) (2021)
- [Optimization Techniques for Unsupervised Complex Table Reasoning via Self-Training Framework](https://mlsanity.com/paper/2212.10097/) (2022)
- [PanelTR: Zero-Shot Table Reasoning Framework Through Multi-Agent Scientific Discussion](https://mlsanity.com/paper/2508.06110/) (2025)

## Abstract (arXiv)

Transformer-based models achieve impressive performance on numerous Natural Language Inference (NLI) benchmarks when trained on respective training datasets. However, in certain cases, training samples may not be available or collecting them could be time-consuming and resource-intensive. In this work, we address the above challenge and present an explorative study on unsupervised NLI, a paradigm in which no human-annotated training samples are available. We investigate it under three settings: PH, P, and NPH that differ in the extent of unlabeled data available for learning. As a solution, we propose a procedural data generation approach that leverages a set of sentence transformations to collect PHL (Premise, Hypothesis, Label) triplets for training NLI models, bypassing the need for human-annotated training data. Comprehensive experiments with several NLI datasets show that the proposed approach results in accuracies of up to 66.75%, 65.9%, 65.39% in PH, P, and NPH settings respectively, outperforming all existing unsupervised baselines. Furthermore, fine-tuning our model with as little as ~0.1% of the human-annotated training dataset (500 instances) leads to 12.2% higher accuracy than the model trained from scratch on the same 500 instances. Supported by this superior performance, we conclude with a recommendation for collecting high-quality task-specific data.