# Mathematical Reasoning on MATH500 (BCE, MSE, Q-value rankings) Source: [Mathematical Reasoning on MATH500 (BCE, MSE, Q-value rankings)]() Collection index: [llms.txt]() Data updated: 2026-07-01T06:48:18.222Z Evaluation of model performance on the MATH500 dataset for Mathematical Reasoning, reporting Binary Cross-Entropy (BCE), Mean Squared Error (MSE), and Q-value rankings. 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 | Mathematical Reasoning | | Dataset (including any stated split) | MATH500 | | Metric | BCE | | Unit | SCALAR | | Direction | Lower is better | | Results loaded | 18 of 18 | | 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 - [BCE](): Lower is better; unit: SCALAR. - [MSE](): Lower is better; unit: SCALAR. - [Q-value Ranking Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | BCE | Result date | Report paper | | --- | --- | --- | --- | | L2R | 18.48 | Not recorded | 68fa7a681e209131cea83008 | | L2R | 18.88 | Not recorded | 68fa7a681e209131cea83008 | | BiPRM | 21 | Not recorded | 68fa7a681e209131cea83008 | | BiPRM | 23.56 | Not recorded | 68fa7a681e209131cea83008 | | L2R | 25.08 | Not recorded | 68fa7a681e209131cea83008 | | L2R | 25.88 | Not recorded | 68fa7a681e209131cea83008 | | L2R | 31.64 | Not recorded | 68fa7a681e209131cea83008 | | BiPRM | 32.64 | Not recorded | 68fa7a681e209131cea83008 | | L2R | 32.76 | Not recorded | 68fa7a681e209131cea83008 | | L2R | 32.8 | Not recorded | 68fa7a681e209131cea83008 | | BiPRM | 33.92 | Not recorded | 68fa7a681e209131cea83008 | | BiPRM | 36.44 | Not recorded | 68fa7a681e209131cea83008 | | BiPRM | 37.32 | Not recorded | 68fa7a681e209131cea83008 | | BiPRM | 38.56 | Not recorded | 68fa7a681e209131cea83008 | | L2R | 40.24 | Not recorded | 68fa7a681e209131cea83008 | | L2R | 41.44 | Not recorded | 68fa7a681e209131cea83008 | | BiPRM | 43.48 | Not recorded | 68fa7a681e209131cea83008 | | BiPRM | 48.48 | Not recorded | 68fa7a681e209131cea83008 | ### L2R BCE: 18.48. Result ID: 6a44b81373d097d845bdb766. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling Policy":"MetaMath-Mistral-7B","Backbone":"Rho-Math-1B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### L2R BCE: 18.88. Result ID: 6a44b81373d097d845bdb76c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling Policy":"Muggle-Math-13B","Backbone":"Rho-Math-1B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### BiPRM BCE: 21. Result ID: 6a44b81373d097d845bdb76d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Policy":"Muggle-Math-13B","Backbone":"Rho-Math-1B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### BiPRM BCE: 23.56. Result ID: 6a44b81373d097d845bdb767. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Policy":"MetaMath-Mistral-7B","Backbone":"Rho-Math-1B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### L2R BCE: 25.08. Result ID: 6a44b81373d097d845bdb770. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling Policy":"Muggle-Math-13B","Backbone":"Deepseek-Math-7B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### L2R BCE: 25.88. Result ID: 6a44b81373d097d845bdb76e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling Policy":"Muggle-Math-13B","Backbone":"Qwen2.5-Math-1.5B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### L2R BCE: 31.64. Result ID: 6a44b81373d097d845bdb76a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling Policy":"MetaMath-Mistral-7B","Backbone":"Deepseek-Math-7B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### BiPRM BCE: 32.64. Result ID: 6a44b81373d097d845bdb771. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Policy":"Muggle-Math-13B","Backbone":"Deepseek-Math-7B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### L2R BCE: 32.76. Result ID: 6a44b81373d097d845bdb772. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling Policy":"Llama-3-70B-Instruct","Backbone":"Rho-Math-1B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### L2R BCE: 32.8. Result ID: 6a44b81373d097d845bdb768. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling Policy":"MetaMath-Mistral-7B","Backbone":"Qwen2.5-Math-1.5B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### BiPRM BCE: 33.92. Result ID: 6a44b81373d097d845bdb76f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Policy":"Muggle-Math-13B","Backbone":"Qwen2.5-Math-1.5B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### BiPRM BCE: 36.44. Result ID: 6a44b81373d097d845bdb76b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Policy":"MetaMath-Mistral-7B","Backbone":"Deepseek-Math-7B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### BiPRM BCE: 37.32. Result ID: 6a44b81373d097d845bdb773. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Policy":"Llama-3-70B-Instruct","Backbone":"Rho-Math-1B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### BiPRM BCE: 38.56. Result ID: 6a44b81373d097d845bdb769. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Policy":"MetaMath-Mistral-7B","Backbone":"Qwen2.5-Math-1.5B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### L2R BCE: 40.24. Result ID: 6a44b81373d097d845bdb776. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling Policy":"Llama-3-70B-Instruct","Backbone":"Deepseek-Math-7B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### L2R BCE: 41.44. Result ID: 6a44b81373d097d845bdb774. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Sampling Policy":"Llama-3-70B-Instruct","Backbone":"Qwen2.5-Math-1.5B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### BiPRM BCE: 43.48. Result ID: 6a44b81373d097d845bdb777. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Policy":"Llama-3-70B-Instruct","Backbone":"Deepseek-Math-7B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. ### BiPRM BCE: 48.48. Result ID: 6a44b81373d097d845bdb775. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampling Policy":"Llama-3-70B-Instruct","Backbone":"Qwen2.5-Math-1.5B","Evaluation Protocol":"Best-of-N (BON@8 to BON@128)"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.