# Evaluating the Robustness of Discrete Prompts

arXiv: https://arxiv.org/abs/2302.05619 · HTML: https://mlsanity.com/paper/2302.05619/

## Key facts

- Cited by 18 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 38 references that are in the map.
- First posted on arXiv 11 Feb 2023.

## Builds on (most cited first, 27 of 27)

- [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)
- [RoBERTa: A Robustly Optimized BERT Pretraining Approach](https://mlsanity.com/paper/1907.11692/) (2019)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Prefix-Tuning: Optimizing Continuous Prompts for Generation](https://mlsanity.com/paper/2101.00190/) (2021)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Neural Machine Translation of Rare Words with Subword Units](https://mlsanity.com/paper/1508.07909/) (2015)
- [A Broad-Coverage Challenge Corpus for Sentence Understanding through Inference](https://mlsanity.com/paper/1704.05426/) (2017)
- [Language Models as Knowledge Bases?](https://mlsanity.com/paper/1909.01066/) (2019)
- [Making Pre-trained Language Models Better Few-shot Learners](https://mlsanity.com/paper/2012.15723/) (2020)
- [Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity](https://mlsanity.com/paper/2104.08786/) (2021)
- [Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference](https://mlsanity.com/paper/2001.07676/) (2020)
- [Universal Adversarial Triggers for Attacking and Analyzing NLP](https://mlsanity.com/paper/1908.07125/) (2019)
- [Is BERT Really Robust? A Strong Baseline for Natural Language Attack on Text Classification and Entailment](https://mlsanity.com/paper/1907.11932/) (2019)
- [RLPrompt: Optimizing Discrete Text Prompts with Reinforcement Learning](https://mlsanity.com/paper/2205.12548/) (2022)
- [Do Prompt-Based Models Really Understand the Meaning of their Prompts?](https://mlsanity.com/paper/2109.01247/) (2021)
- [On the Stability of Fine-tuning BERT: Misconceptions, Explanations, and Strong Baselines](https://mlsanity.com/paper/2006.04884/) (2020)
- [Revisiting Few-sample BERT Fine-tuning](https://mlsanity.com/paper/2006.05987/) (2020)
- [WARP: Word-level Adversarial ReProgramming](https://mlsanity.com/paper/2101.00121/) (2021)
- [How Many Data Points is a Prompt Worth?](https://mlsanity.com/paper/2103.08493/) (2021)
- [Template-Based Named Entity Recognition Using BART](https://mlsanity.com/paper/2106.01760/) (2021)
- [Towards Crafting Text Adversarial Samples](https://mlsanity.com/paper/1707.02812/) (2017)
- [True Few-Shot Learning with Prompts -- A Real-World Perspective](https://mlsanity.com/paper/2111.13440/) (2021)
- [Evaluating Gender Bias in Natural Language Inference](https://mlsanity.com/paper/2105.05541/) (2021)
- [Avoiding Inference Heuristics in Few-shot Prompt-based Finetuning](https://mlsanity.com/paper/2109.04144/) (2021)
- [Are Prompt-based Models Clueless?](https://mlsanity.com/paper/2205.09295/) (2022)
- [Signal Processing on PV Time-Series Data: Robust Degradation Analysis without Physical Models](https://mlsanity.com/paper/1907.09456/) (2019)

## Built on by (most cited first, 18 of 18)

- [Value FULCRA: Mapping Large Language Models to the Multidimensional Spectrum of Basic Human Values](https://mlsanity.com/paper/2311.10766/) (2023)
- [The language of prompting: What linguistic properties make a prompt successful?](https://mlsanity.com/paper/2311.01967/) (2023)
- [Large Language Models as Tax Attorneys: A Case Study in Legal Capabilities Emergence](https://mlsanity.com/paper/2306.07075/) (2023)
- [Give Me the Facts! A Survey on Factual Knowledge Probing in Pre-trained Language Models](https://mlsanity.com/paper/2310.16570/) (2023)
- [Denevil: Towards Deciphering and Navigating the Ethical Values of Large Language Models via Instruction Learning](https://mlsanity.com/paper/2310.11053/) (2023)
- [InstOptima: Evolutionary Multi-objective Instruction Optimization via Large Language Model-based Instruction Operators](https://mlsanity.com/paper/2310.17630/) (2023)
- [Empowering Large Language Models for Textual Data Augmentation](https://mlsanity.com/paper/2404.17642/) (2024)
- [Benchmarking and Explaining Large Language Model-based Code Generation: A Causality-Centric Approach](https://mlsanity.com/paper/2310.06680/) (2023)
- [Unnatural language processing: How do language models handle machine-generated prompts?](https://mlsanity.com/paper/2310.15829/) (2023)
- [Brittle Minds, Fixable Activations: Understanding Belief Representations in Language Models](https://mlsanity.com/paper/2406.17513/) (2024)
- [Test-Time Self-Adaptive Small Language Models for Question Answering](https://mlsanity.com/paper/2310.13307/) (2023)
- [Prompts have evil twins](https://mlsanity.com/paper/2311.07064/) (2023)
- [Evil twins are not that evil: Qualitative insights into machine-generated prompts](https://mlsanity.com/paper/2412.08127/) (2024)
- [JevOut: Natural Context Can Flip Decision Models](https://mlsanity.com/paper/2609.30243/) (2026)
- [PAFT: Prompt-Agnostic Fine-Tuning](https://mlsanity.com/paper/2502.12859/) (2025)
- [Demystifying optimized prompts in language models](https://mlsanity.com/paper/2505.02273/) (2025)
- [POaaS: Minimal-Edit Prompt Optimization as a Service to Lift Accuracy and Cut Hallucinations on On-Device sLLMs](https://mlsanity.com/paper/2603.16045/) (2026)
- [Prompts Without Evidence: How Neuroimaging Mentions Shift Clinical Vision-Language Model Predictions](https://mlsanity.com/paper/2603.28387/) (2026)

## Abstract (arXiv)

Discrete prompts have been used for fine-tuning Pre-trained Language Models for diverse NLP tasks. In particular, automatic methods that generate discrete prompts from a small set of training instances have reported superior performance. However, a closer look at the learnt prompts reveals that they contain noisy and counter-intuitive lexical constructs that would not be encountered in manually-written prompts. This raises an important yet understudied question regarding the robustness of automatically learnt discrete prompts when used in downstream tasks. To address this question, we conduct a systematic study of the robustness of discrete prompts by applying carefully designed perturbations into an application using AutoPrompt and then measure their performance in two Natural Language Inference (NLI) datasets. Our experimental results show that although the discrete prompt-based method remains relatively robust against perturbations to NLI inputs, they are highly sensitive to other types of perturbations such as shuffling and deletion of prompt tokens. Moreover, they generalize poorly across different NLI datasets. We hope our findings will inspire future work on robust discrete prompt learning.