# From Feelings to Metrics: Understanding and Formalizing How Users Vibe-Test LLMs

arXiv: https://arxiv.org/abs/2604.14137 · HTML: https://mlsanity.com/paper/2604.14137/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 36 references that are in the map.
- First posted on arXiv 15 Apr 2026.

## Builds on (most cited first, 26 of 26)

- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code](https://mlsanity.com/paper/2403.07974/) (2024)
- [Is Your Code Generated by ChatGPT Really Correct? Rigorous Evaluation of Large Language Models for Code Generation](https://mlsanity.com/paper/2305.01210/) (2023)
- [gpt-oss-120b & gpt-oss-20b Model Card](https://mlsanity.com/paper/2508.10925/) (2025)
- [Know Me, Respond to Me: Benchmarking LLMs for Dynamic User Profiling and Personalized Responses at Scale](https://mlsanity.com/paper/2504.14225/) (2025)
- [DataPerf: Benchmarks for Data-Centric AI Development](https://mlsanity.com/paper/2207.10062/) (2022)
- [A Systematic Survey and Critical Review on Evaluating Large Language Models: Challenges, Limitations, and Recommendations](https://mlsanity.com/paper/2407.04069/) (2024)
- [EvalLM: Interactive Evaluation of Large Language Model Prompts on User-Defined Criteria](https://mlsanity.com/paper/2309.13633/) (2023)
- [PersonaLens: A Benchmark for Personalization Evaluation in Conversational AI Assistants](https://mlsanity.com/paper/2506.09902/) (2025)
- [Toward an Evaluation Science for Generative AI Systems](https://mlsanity.com/paper/2503.05336/) (2025)
- [Know You First and Be You Better: Modeling Human-Like User Simulators via Implicit Profiles](https://mlsanity.com/paper/2502.18968/) (2025)
- [Automatic Interactive Evaluation for Large Language Models with State Aware Patient Simulator](https://mlsanity.com/paper/2403.08495/) (2024)
- [ChatBench: From Static Benchmarks to Human-AI Evaluation](https://mlsanity.com/paper/2504.07114/) (2025)
- [LMUnit: Fine-grained Evaluation with Natural Language Unit Tests](https://mlsanity.com/paper/2412.13091/) (2024)
- [CodeJudgeBench: Benchmarking LLM-as-a-Judge for Coding Tasks](https://mlsanity.com/paper/2507.10535/) (2025)
- [AMONGAGENTS: Evaluating Large Language Models in the Interactive Text-Based Social Deduction Game](https://mlsanity.com/paper/2407.16521/) (2024)
- [Can LLM Agents Simulate Multi-Turn Human Behavior? Evidence from Real Online Customer Behavior Data](https://mlsanity.com/paper/2503.20749/) (2025)
- [Toward Generalizable Evaluation in the LLM Era: A Survey Beyond Benchmarks](https://mlsanity.com/paper/2504.18838/) (2025)
- [DuetSim: Building User Simulator with Dual Large Language Models for Task-Oriented Dialogues](https://mlsanity.com/paper/2405.13028/) (2024)
- [User Simulation with Large Language Models for Evaluating Task-Oriented Dialogue](https://mlsanity.com/paper/2309.13233/) (2023)
- [IQA-EVAL: Automatic Evaluation of Human-Model Interactive Question Answering](https://mlsanity.com/paper/2408.13545/) (2024)
- [VibeCheck: Discover and Quantify Qualitative Differences in Large Language Models](https://mlsanity.com/paper/2410.12851/) (2024)
- [EvalAgent: Discovering Implicit Evaluation Criteria from the Web](https://mlsanity.com/paper/2504.15219/) (2025)
- [SimulBench: Evaluating Language Models with Creative Simulation Tasks](https://mlsanity.com/paper/2409.07641/) (2024)
- [PREF: Reference-Free Evaluation of Personalised Text Generation in LLMs](https://mlsanity.com/paper/2508.10028/) (2025)
- [PromptSuite: A Task-Agnostic Framework for Multi-Prompt Generation](https://mlsanity.com/paper/2507.14913/) (2025)

## Abstract (arXiv)

Evaluating LLMs is challenging, as benchmark scores often fail to capture models' real-world usefulness. Instead, users often rely on ``vibe-testing'': informal experience-based evaluation, such as comparing models on coding tasks related to their own workflow. While prevalent, vibe-testing is often too ad hoc and unstructured to analyze or reproduce at scale. In this work, we study how vibe-testing works in practice and then formalize it to support systematic analysis. We first analyze two empirical resources: (1) a survey of user evaluation practices, and (2) a collection of in-the-wild model comparison reports from blogs and social media. Based on these resources, we formalize vibe-testing as a two-part process: users personalize both what they test and how they judge responses. We then introduce a proof-of-concept evaluation pipeline that follows this formulation by generating personalized prompts and comparing model outputs using user-aware subjective criteria. In experiments on coding benchmarks, we find that combining personalized prompts and user-aware evaluation can change which model is preferred, reflecting the role of vibe-testing in practice. These findings suggest that formalized vibe-testing can serve as a useful approach for bridging benchmark scores and real-world experience.