# LLM Judge Evaluation on LLM-to-LLM Evaluation Reference: GPT-5.2

Source: [LLM Judge Evaluation on LLM-to-LLM Evaluation Reference: GPT-5.2](<https://www.sota2.com/research/sota/llm-judge-evaluation-on-llm-to-llm-evaluation-reference-gpt-5-2>)

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

Data updated: 2026-03-16T08:50:16.058Z

Evaluation of LLM performance using the LLM Judge methodology against the GPT-5.2 reference dataset, reporting correlation (r), gap, and recovery metrics.

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 | LLM Judge Evaluation |
| Dataset (including any stated split) | LLM-to-LLM Evaluation Reference: GPT-5.2 |
| Metric | Global Correlation (r) |
| Unit | SCALAR |
| 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

- [Global Correlation (r)](<https://www.sota2.com/research/sota/llm-judge-evaluation-on-llm-to-llm-evaluation-reference-gpt-5-2.md?metric=Global%20Correlation%20(r)>): Higher is better; unit: SCALAR.

- [Within-Group Correlation (r)](<https://www.sota2.com/research/sota/llm-judge-evaluation-on-llm-to-llm-evaluation-reference-gpt-5-2.md?metric=Within-Group%20Correlation%20(r)>): Higher is better; unit: SCALAR.

- [Performance Gap (%)](<https://www.sota2.com/research/sota/llm-judge-evaluation-on-llm-to-llm-evaluation-reference-gpt-5-2.md?metric=Performance%20Gap%20(%25)>): Lower is better; unit: PERCENT.

- [Performance Recovery (%)](<https://www.sota2.com/research/sota/llm-judge-evaluation-on-llm-to-llm-evaluation-reference-gpt-5-2.md?metric=Performance%20Recovery%20(%25)>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Global Correlation (r) | Result date | Report paper |
| --- | --- | --- | --- |
| GPT-5-mini | 0.84 | Not recorded | 69b78e9e0636e5bf4afbdcc9 |
| GPT-4.1-nano | 0.49 | Not recorded | 69b78e9e0636e5bf4afbdcc9 |

### GPT-5-mini

Global Correlation (r): 0.84. Result ID: 69b7c43147f34ce9249f7f83.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e9e0636e5bf4afbdcc9.md>)

Recorded evaluation configuration: {"Relative Cost":"7x cheaper"}. This record does not assert independent reproduction.

### GPT-4.1-nano

Global Correlation (r): 0.49. Result ID: 69b7c43147f34ce9249f7f84.

[Reporting paper](<https://www.sota2.com/research/paper/69b78e9e0636e5bf4afbdcc9.md>)

Recorded evaluation configuration: {"Relative Cost":"35x cheaper"}. This record does not assert independent reproduction.

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