# From Subjective Judgments to Auditable Standards:Protocol-Guided AI Auditing of Website Redundancy

arXiv: https://arxiv.org/abs/2608.21476 · HTML: https://mlsanity.com/paper/2608.21476/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 21 of its 38 references that are in the map.
- First posted on arXiv 21 Aug 2026.

## Builds on (most cited first, 21 of 21)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [Evaluating Object Hallucination in Large Vision-Language Models](https://mlsanity.com/paper/2305.10355/) (2023)
- [Large Language Models are not Fair Evaluators](https://mlsanity.com/paper/2305.17926/) (2023)
- [Model Cards for Model Reporting](https://mlsanity.com/paper/1810.03993/) (2018)
- [Beyond Accuracy: Behavioral Testing of NLP models with CheckList](https://mlsanity.com/paper/2005.04118/) (2020)
- [Selective Classification for Deep Neural Networks](https://mlsanity.com/paper/1705.08500/) (2017)
- [Enabling Large Language Models to Generate Text with Citations](https://mlsanity.com/paper/2305.14627/) (2023)
- [SeeClick: Harnessing GUI Grounding for Advanced Visual GUI Agents](https://mlsanity.com/paper/2401.10935/) (2024)
- [Improving Reproducibility in Machine Learning Research (A Report from the NeurIPS 2019 Reproducibility Program)](https://mlsanity.com/paper/2003.12206/) (2020)
- [Measurement and Fairness](https://mlsanity.com/paper/1912.05511/) (2019)
- [Measuring Attribution in Natural Language Generation Models](https://mlsanity.com/paper/2112.12870/) (2021)
- [Benchmarking Cognitive Biases in Large Language Models as Evaluators](https://mlsanity.com/paper/2309.17012/) (2023)
- [ScreenAI: A Vision-Language Model for UI and Infographics Understanding](https://mlsanity.com/paper/2402.04615/) (2024)
- [MLLM as a UI Judge: Benchmarking Multimodal LLMs for Predicting Human Perception of User Interfaces](https://mlsanity.com/paper/2510.08783/) (2025)
- [Code Aesthetics with Agentic Reward Feedback](https://mlsanity.com/paper/2510.23272/) (2025)
- [UXCascade: Scalable Usability Testing with Simulated User Agents](https://mlsanity.com/paper/2601.15777/) (2026)
- [Do MLLMs Capture How Interfaces Guide User Behavior? A Benchmark for Multimodal UI/UX Design Understanding](https://mlsanity.com/paper/2505.05026/) (2025)
- [Avenir-UX: Automated UX Evaluation via Simulated Human Web Interaction with GUI Grounding](https://mlsanity.com/paper/2604.09581/) (2026)
- [Training Computer Use Agents to Assess the Usability of Graphical User Interfaces](https://mlsanity.com/paper/2604.26020/) (2026)
- [Unraveling the Impact of Visual Complexity on Search as Learning](https://mlsanity.com/paper/2501.05289/) (2025)

## Abstract (arXiv)

Website redundancy does not have a single fixed meaning. The same repeated element may distract during one task and provide backup during another. We introduce CORA (Counterfactual, Observable Redundancy Audit), which measures repetition load, normal-use tax, and failure-domain recovery reserve separately. Each run retains screenshots, stable element identities, and task traces. A versioned vision-language model proposes the annotations. Typed validation and release checks then determine whether a calibrated dimension can be reported; failed or malformed outputs stay in the fixed denominator. On a transparent mechanistic testbed, the factorized CORA representation separated reserve from normal-use tax and predicted perturbed success more accurately than scalar-load baselines. The model studies then showed why repeatability is not enough: two small local vision-language models produced recurring outputs, but neither instrument met all release requirements. CORA therefore withheld automated scores from both instruments while retaining the raw responses and failure records. Separate checker fixtures confirmed that the typed validator and hardened release gates implement their specifications; these tests do not establish semantic grounding or accuracy on production sites. Taken together, the results position CORA as an auditable candidate procedure for the controlled benchmark studied here rather than a general standard. Human agreement, AI-versus-human accuracy, and validation on independent production sites remain open empirical questions.