# Brittlebench: Quantifying LLM robustness via prompt sensitivity

arXiv: https://arxiv.org/abs/2603.13285 · HTML: https://mlsanity.com/paper/2603.13285/

## Key facts

- Cited by 21 papers in the map (data through 14 Oct 2026).
- Builds on 57 of its 65 references that are in the map.
- First posted on arXiv 27 Feb 2026.

## Builds on (most cited first, 30 of 57)

- [GPT-4 Technical Report](https://mlsanity.com/paper/2303.08774/) (2023)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Measuring Massive Multitask Language Understanding](https://mlsanity.com/paper/2009.03300/) (2020)
- [Think you have Solved Question Answering? Try ARC, the AI2 Reasoning Challenge](https://mlsanity.com/paper/1803.05457/) (2018)
- [TruthfulQA: Measuring How Models Mimic Human Falsehoods](https://mlsanity.com/paper/2109.07958/) (2021)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [Beyond the Imitation Game: Quantifying and extrapolating the capabilities of language models](https://mlsanity.com/paper/2206.04615/) (2022)
- [Shortcut Learning in Deep Neural Networks](https://mlsanity.com/paper/2004.07780/) (2020)
- [Qwen3 Embedding: Advancing Text Embedding and Reranking Through Foundation Models](https://mlsanity.com/paper/2506.05176/) (2025)
- [Chatbot Arena: An Open Platform for Evaluating LLMs by Human Preference](https://mlsanity.com/paper/2403.04132/) (2024)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Right for the Wrong Reasons: Diagnosing Syntactic Heuristics in Natural Language Inference](https://mlsanity.com/paper/1902.01007/) (2019)
- [Quantifying Language Models' Sensitivity to Spurious Features in Prompt Design or: How I learned to start worrying about prompt formatting](https://mlsanity.com/paper/2310.11324/) (2023)
- [LLM Evaluators Recognize and Favor Their Own Generations](https://mlsanity.com/paper/2404.13076/) (2024)
- [Annotation Artifacts in Natural Language Inference Data](https://mlsanity.com/paper/1803.02324/) (2018)
- [MathQA: Towards Interpretable Math Word Problem Solving with Operation-Based Formalisms](https://mlsanity.com/paper/1905.13319/) (2019)
- [Dynabench: Rethinking Benchmarking in NLP](https://mlsanity.com/paper/2104.14337/) (2021)
- [AI and the Everything in the Whole Wide World Benchmark](https://mlsanity.com/paper/2111.15366/) (2021)
- [State of What Art? A Call for Multi-Prompt LLM Evaluation](https://mlsanity.com/paper/2401.00595/) (2023)
- [Synthetic and Natural Noise Both Break Neural Machine Translation](https://mlsanity.com/paper/1711.02173/) (2017)
- [Large Language Models Sensitivity to The Order of Options in Multiple-Choice Questions](https://mlsanity.com/paper/2308.11483/) (2023)
- [PromptRobust: Towards Evaluating the Robustness of Large Language Models on Adversarial Prompts](https://mlsanity.com/paper/2306.04528/) (2023)
- [What Will it Take to Fix Benchmarking in Natural Language Understanding?](https://mlsanity.com/paper/2104.02145/) (2021)
- [Does Writing with Language Models Reduce Content Diversity?](https://mlsanity.com/paper/2309.05196/) (2023)
- [ProSA: Assessing and Understanding the Prompt Sensitivity of LLMs](https://mlsanity.com/paper/2410.12405/) (2024)
- [Large Language Models Understand and Can be Enhanced by Emotional Stimuli](https://mlsanity.com/paper/2307.11760/) (2023)
- [Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselves](https://mlsanity.com/paper/2311.04205/) (2023)
- [Measuring what Matters: Construct Validity in Large Language Model Benchmarks](https://mlsanity.com/paper/2511.04703/) (2025)

## Built on by (most cited first, 21 of 21)

- [Hidden Measurement Error in LLM Pipelines Distorts Annotation, Evaluation, and Benchmarking](https://mlsanity.com/paper/2604.11581/) (2026)
- [JudgeSense: A Benchmark for Prompt Sensitivity in LLM-as-a-Judge Systems](https://mlsanity.com/paper/2604.23478/) (2026)
- [Covert Influence Between Language Models](https://mlsanity.com/paper/2606.04071/) (2026)
- [One prompt is not enough: Instruction Sensitivity Undermines Embedding Model Evaluation](https://mlsanity.com/paper/2605.22544/) (2026)
- [Efficient Benchmarking Is Just Feature Selection and Multiple Regression](https://mlsanity.com/paper/2605.25773/) (2026)
- [Online Shift Detection and Conformal Adaptation for Deployed Safety Classifiers](https://mlsanity.com/paper/2606.11949/) (2026)
- [SCRuB: Social Concept Reasoning under Rubric-Based Evaluation](https://mlsanity.com/paper/2605.06444/) (2026)
- [PRISM: A Benchmark for Programmatic Spatial-Temporal Reasoning](https://mlsanity.com/paper/2605.19382/) (2026)
- [Are We Evaluating Knowledge or Phrasing? Mitigating MCQA Sensitivity with ParaEval](https://mlsanity.com/paper/2606.10657/) (2026)
- [Same Evidence, Different Answer: Auditing Order Sensitivity in Multimodal Large Language Models](https://mlsanity.com/paper/2606.26079/) (2026)
- [Jagged Judges: Epistemic Stability Under Perturbation, Pressure, and Persistence](https://mlsanity.com/paper/2608.12645/) (2026)
- [In-Situ Behavioral Evaluation for LLM Fairness, Not Standardized-Test Scores](https://mlsanity.com/paper/2605.12530/) (2026)
- [Does Slightly Mean Somewhat? Measuring Vague Intensity Words in LLM Numeric Actions](https://mlsanity.com/paper/2605.21827/) (2026)
- [Greener Than Humans? Environmental Attitudes in Large Language Models](https://mlsanity.com/paper/2606.02741/) (2026)
- [A Definition of Good Explanations and the Challenges Explaining LLM Outputs](https://mlsanity.com/paper/2606.14838/) (2026)
- [Surrogate Fidelity: When Can Open LLMs Explain Closed Ones?](https://mlsanity.com/paper/2606.32008/) (2026)
- [Thought without systematicity? Evaluating reasoning models on rule induction tasks](https://mlsanity.com/paper/2609.13948/) (2026)
- [CHART: A Harness-Rotation Curriculum for Harness-Robust Search Agents](https://mlsanity.com/paper/2609.22247/) (2026)
- [Latent Inference-Time Guidance of Time Series Foundation Models](https://mlsanity.com/paper/2609.38058/) (2026)
- [Quantization Effects on Tool-Failure Recovery Vary Across Prompts and Evaluation Designs](https://mlsanity.com/paper/2610.07781/) (2026)
- [perturber: A Stateless Service for Semantics-Preserving Text Perturbations](https://mlsanity.com/paper/2610.12667/) (2026)

## Abstract (arXiv)

Existing evaluation methods largely rely on clean, static benchmarks, which can overestimate true model performance by failing to capture the noise and variability inherent in real-world user inputs. This is especially true for language models, which can face human-generated text queries containing mistakes, typos, or alternative ways of phrasing the same question. In this work, we introduce a theoretical framework for quantifying model sensitivity to prompt variants, or brittleness, that can enable us to disentangle data-induced difficulty from prompt-related variability. Using this framework, we design a novel evaluation pipeline, Brittlebench, to holistically evaluate the sensitivity of frontier models. We apply semantics-preserving perturbations to a suite of popular benchmarks, and observe model performance to degrade as much as 12%. However, these perturbations do not affect all models equally: even a single perturbation alters the relative ranking of models in 63% of cases, impacting conclusions about comparative model performance. Decomposing the total variance of both state-of-the-art open-weight and commercial models, we find that semantics-preserving input perturbations can account for up to half of the performance variance for a given model. Brittlebench highlights the need for more robust evaluations and models, and allows us to systematically understand model brittleness.