# Scaling Model Validation on BrowseComp-Plus Out-of-sample (val)

Source: [Scaling Model Validation on BrowseComp-Plus Out-of-sample (val)](<https://www.sota2.com/research/sota/scaling-model-validation-on-browsecomp-plus-out-of-sample-val>)

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

Data updated: 2026-02-26T05:49:09.839Z

Evaluation of model performance on Scaling Model Validation using the out-of-sample validation split of the BrowseComp-Plus dataset, measured using metrics like MAE, MAPE, and Kendall’s τ.

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 | Scaling Model Validation |
| Dataset (including any stated split) | BrowseComp-Plus Out-of-sample (val) |
| Metric | MAE |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 1 of 1 |
| 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

- [MAE](<https://www.sota2.com/research/sota/scaling-model-validation-on-browsecomp-plus-out-of-sample-val.md?metric=MAE>): Lower is better; unit: SCALAR.

- [MAPE](<https://www.sota2.com/research/sota/scaling-model-validation-on-browsecomp-plus-out-of-sample-val.md?metric=MAPE>): Lower is better; unit: PERCENT.

- [Normalized MAE](<https://www.sota2.com/research/sota/scaling-model-validation-on-browsecomp-plus-out-of-sample-val.md?metric=Normalized%20MAE>): Lower is better; unit: SCALAR.

- [Qualitative Validation Score](<https://www.sota2.com/research/sota/scaling-model-validation-on-browsecomp-plus-out-of-sample-val.md?metric=Qualitative%20Validation%20Score>): Higher is better; unit: PERCENT.

- [Kendall's τ (Ranking)](<https://www.sota2.com/research/sota/scaling-model-validation-on-browsecomp-plus-out-of-sample-val.md?metric=Kendall%27s%20%CF%84%20(Ranking)>): Higher is better; unit: SCALAR.

## Recorded results

| Method | MAE | Result date | Report paper |
| --- | --- | --- | --- |
| predictive framework for agentic scaling | 0.071 | Not recorded | 6938fe796cfd488761a53b9a |

### predictive framework for agentic scaling

MAE: 0.071. Result ID: 69962c105698665ec46d691d.

[Reporting paper](<https://www.sota2.com/research/paper/6938fe796cfd488761a53b9a.md>)

[Method origin paper](<https://www.sota2.com/research/paper/6938fe796cfd488761a53b9a.md>)

Recorded evaluation configuration: {"Model":"GPT-5.2","Intelligence Index":"75"}. This record does not assert independent reproduction.

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