Research map / Vision-language and multimodal / Human and embodied modeling
The Stanford Acuity Test: A Precise Vision Test Using Bayesian Techniques and a Discovery in Human Visual Response
Chris Piech, Ali Malik, Laura M. Scott, Robert T. Chang, Charles Lin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Use StAT instead of Snellen | visual acuity testing · Simulated patients | average relative error: 0.264 -> 0.069 | Improved |
| Use StAT instead of FrACT | visual acuity testing · Simulated patients, 20 questions | average relative error: 0.212 -> 0.069 | Improved |
| Add slip-probability modeling (ablation) | StAT · Simulated patients, 20 questions | average relative error: 0.150 -> 0.069 | Improved |
| Use posterior sampling instead of greedy MAP query selection (ablation) | StAT · Simulated patients, 20 questions | average relative error: 0.132 -> 0.069 | Improved |
| Use the Floored Exponential VRF instead of a logistic VRF (ablation) | StAT · Simulated patients, 20 questions | average relative error: 0.125 -> 0.069 | Improved |
| Add an acuity prior (ablation) | StAT · Simulated patients, 20 questions | average relative error: 0.090 -> 0.069 | Improved |
| Use an informed prior | StAT · Simulated patients, 20 questions | average relative error: 0.069 -> 0.047 | Improved |
| Use a confidence-based stopping rule | StAT · Simulated patients | average relative error: 0.069 -> 0.038 | Improved |
| Use a confidence-based stopping rule | StAT · Simulated patients | average test length: 20 -> 63 | Hurt |
| Increase exam length to 200 letters | StAT · Simulated patients | average relative error: 0.069 -> 0.020 | Improved |
| Use 19-letter optotype options rather than four | StAT · Simulated patients | average relative error: 0.069 -> 0.052 | Improved |
| Increase slip probability while it remains below 1/3 | StAT · Simulated patients | average relative error | No clear effect |
Closest papers by meaning
- RLPeri: Accelerating Visual Perimetry Test with Reinforcement Learning and Convolutional Feature Extraction (AAAI 2024 · similarity 0.85)
- FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMs (NeurIPS 2026 · similarity 0.82)
- A World Model of Radiologist Reading for Medical Image Representation Learning (NeurIPS 2026 · similarity 0.82)
- Statistical Test for Attention Maps in Vision Transformers (ICML 2024 · similarity 0.82)
- Med-Scout: Curing MLLMs' Geometric Blindness in Medical Perception via Geometry-Aware RL Post-Training (ICML 2026 · similarity 0.80)
- Visual and Domain Knowledge for Professional-level Graph-of-Thought Medical Reasoning (ICML 2025 · similarity 0.80)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
