# Unseen prompt generalization on ImageNet

Source: [Unseen prompt generalization on ImageNet](<https://www.sota2.com/research/sota/unseen-prompt-generalization-on-imagenet>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:29:02.805Z

Evaluation of model performance on the ImageNet dataset specifically testing generalization capabilities using unseen prompts, measured by Accuracy.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Unseen prompt generalization |
| Dataset (including any stated split) | ImageNet |
| Metric | Accuracy |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 2 of 2 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Accuracy](<https://www.sota2.com/research/sota/unseen-prompt-generalization-on-imagenet.md?metric=Accuracy>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Accuracy | Result date | Report paper |
| --- | --- | --- | --- |
| Bayesian Prompt Learning | 70.93 | Not recorded | 68fa70571e209131cea4a691 |
| CLIP | 68.14 | Not recorded | 68fa70571e209131cea4a691 |

### Bayesian Prompt Learning

Accuracy: 70.93. Result ID: 699114e73a6c314849058946.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70571e209131cea4a691.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa70571e209131cea4a691.md>)

[Implementation](<https://github.com/saic-fi/Bayesian-Prompt-Learning>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Prompt Length":4,"Number of Shots":16,"Monte Carlo samples (MC)":10,"Epochs":10}. This record does not assert independent reproduction.

### CLIP

Accuracy: 68.14. Result ID: 699114e73a6c314849058945.

[Reporting paper](<https://www.sota2.com/research/paper/68fa70571e209131cea4a691.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa69581e209131cea21ad0.md>)

[Implementation](<https://github.com/OpenAI/CLIP>)

Recorded evaluation configuration: {"Backbone":"ViT-B/16","Prompt":"Fixed-prompt"}. This record does not assert independent reproduction.

Page 1 of 1. This is an index page, not the full collection.