# Unsupervised Label Refinement Improves Dataless Text Classification

arXiv: https://arxiv.org/abs/2012.04194 · HTML: https://mlsanity.com/paper/2012.04194/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 8 Dec 2020.

## Built on by (most cited first, 9 of 9)

- [True Few-Shot Learning with Prompts -- A Real-World Perspective](https://mlsanity.com/paper/2111.13440/) (2021)
- [Few-Shot Learning with Siamese Networks and Label Tuning](https://mlsanity.com/paper/2203.14655/) (2022)
- [Beyond Prompting: Making Pre-trained Language Models Better Zero-shot Learners by Clustering Representations](https://mlsanity.com/paper/2210.16637/) (2022)
- [The Benefits of Label-Description Training for Zero-Shot Text Classification](https://mlsanity.com/paper/2305.02239/) (2023)
- [Zero and Few-shot Learning for Author Profiling](https://mlsanity.com/paper/2204.10543/) (2022)
- [DocSCAN: Unsupervised Text Classification via Learning from Neighbors](https://mlsanity.com/paper/2105.04024/) (2021)
- [Unsupervised Ranking and Aggregation of Label Descriptions for Zero-Shot Classifiers](https://mlsanity.com/paper/2204.09481/) (2022)
- [Exploring Description-Augmented Dataless Intent Classification](https://mlsanity.com/paper/2407.17862/) (2024)
- [Soft Contextualized Encoder For User Defined Text Classification](https://mlsanity.com/paper/2601.03450/) (2026)

## Abstract (arXiv)

Dataless text classification is capable of classifying documents into previously unseen labels by assigning a score to any document paired with a label description. While promising, it crucially relies on accurate descriptions of the label set for each downstream task. This reliance causes dataless classifiers to be highly sensitive to the choice of label descriptions and hinders the broader application of dataless classification in practice. In this paper, we ask the following question: how can we improve dataless text classification using the inputs of the downstream task dataset? Our primary solution is a clustering based approach. Given a dataless classifier, our approach refines its set of predictions using k-means clustering. We demonstrate the broad applicability of our approach by improving the performance of two widely used classifier architectures, one that encodes text-category pairs with two independent encoders and one with a single joint encoder. Experiments show that our approach consistently improves dataless classification across different datasets and makes the classifier more robust to the choice of label descriptions.