# Model Ranking Correlation Analysis on SEED-Bench (10%-50% subsets) Source: [Model Ranking Correlation Analysis on SEED-Bench (10%-50% subsets)]() Collection index: [llms.txt]() Data updated: 2026-03-04T19:39:18.141Z Evaluation of model ranking correlation performance on the SEED-Bench dataset, specifically measuring Spearman's rank correlation across various data subset sizes (10% through 50%). 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 | Model Ranking Correlation Analysis | | Dataset (including any stated split) | SEED-Bench | | Metric | Spearman Correlation (10% subset) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 7 of 7 | | 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 - [Spearman Correlation (10% subset)](): Higher is better; unit: SCALAR. - [Spearman Correlation (20% subset)](): Higher is better; unit: SCALAR. - [Spearman Correlation (30% subset)](): Higher is better; unit: SCALAR. - [Spearman Correlation (40% subset)](): Higher is better; unit: SCALAR. - [Spearman Correlation (50% subset)](): Higher is better; unit: SCALAR. ## Recorded results | Method | Spearman Correlation (10% subset) | Result date | Report paper | | --- | --- | --- | --- | | M2IRT | 0.94 | Not recorded | 69a7bc9a2d603e6de3e3d6a5 | | M3IRT | 0.94 | Not recorded | 69a7bc9a2d603e6de3e3d6a5 | | FlashEval | 0.89 | Not recorded | 69a7bc9a2d603e6de3e3d6a5 | | IRT | 0.86 | Not recorded | 69a7bc9a2d603e6de3e3d6a5 | | Random | 0.86 | Not recorded | 69a7bc9a2d603e6de3e3d6a5 | | TinyBenchmarks | 0.69 | Not recorded | 69a7bc9a2d603e6de3e3d6a5 | | MIRT | 0.42 | Not recorded | 69a7bc9a2d603e6de3e3d6a5 | ### M2IRT Spearman Correlation (10% subset): 0.94. Result ID: 69a88a4a6dbb7075b9510e25. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### M3IRT Spearman Correlation (10% subset): 0.94. Result ID: 69a88a4a6dbb7075b9510e26. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### FlashEval Spearman Correlation (10% subset): 0.89. Result ID: 69a88a4a6dbb7075b9510e2a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### IRT Spearman Correlation (10% subset): 0.86. Result ID: 69a88a4a6dbb7075b9510e27. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Random Spearman Correlation (10% subset): 0.86. Result ID: 69a88a4a6dbb7075b9510e2b. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### TinyBenchmarks Spearman Correlation (10% subset): 0.69. Result ID: 69a88a4a6dbb7075b9510e29. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### MIRT Spearman Correlation (10% subset): 0.42. Result ID: 69a88a4a6dbb7075b9510e28. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.