# Religious Bias Evaluation on Multilingual CrowS-Pairs (test) Source: [Religious Bias Evaluation on Multilingual CrowS-Pairs (test)]() Collection index: [llms.txt]() Data updated: 2026-04-06T08:14:26.707Z Evaluation of religious bias using the Multilingual CrowS-Pairs dataset, reporting language-specific scores (DE, ZH, ES, JA) on the test split. 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 | Religious Bias Evaluation | | Dataset (including any stated split) | Multilingual CrowS-Pairs (test) | | Metric | Bias Score (DE) | | Unit | PERCENT | | 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 - [Bias Score (DE)](): Lower is better; unit: PERCENT. - [Bias Score (ZH)](): Lower is better; unit: PERCENT. - [Bias Score (ES)](): Lower is better; unit: PERCENT. - [Bias Score (JA)](): Lower is better; unit: PERCENT. - [Average Bias Score](): Lower is better; unit: PERCENT. ## Recorded results | Method | Bias Score (DE) | Result date | Report paper | | --- | --- | --- | --- | | XLM-R + MD w/ Full Fine-Tune | 4.17 | Not recorded | 69d33dd97d0172140ec734ee | | mBERT + SD | 6.67 | Not recorded | 69d33dd97d0172140ec734ee | | XLM-R + MD w/ Prefix Tune | 7.5 | Not recorded | 69d33dd97d0172140ec734ee | | XLM-R + MD w/ Adapter Tune | 9.17 | Not recorded | 69d33dd97d0172140ec734ee | | XLM-R + MCDA | 11.88 | Not recorded | 69d33dd97d0172140ec734ee | | XLM-R + MD w/ Prompt Tune | 13.33 | Not recorded | 69d33dd97d0172140ec734ee | | mBERT + MSD | 15.83 | Not recorded | 69d33dd97d0172140ec734ee | | XLM-R + CDA | 16.67 | Not recorded | 69d33dd97d0172140ec734ee | | XLM-R + SD | 16.67 | Not recorded | 69d33dd97d0172140ec734ee | | mBERT + MCDA | 17.71 | Not recorded | 69d33dd97d0172140ec734ee | | mBERT | 19.17 | Not recorded | 69d33dd97d0172140ec734ee | | XLM-R + MSD | 20 | Not recorded | 69d33dd97d0172140ec734ee | | XLM-R | 21.67 | Not recorded | 69d33dd97d0172140ec734ee | | mBERT + CDA | 22.5 | Not recorded | 69d33dd97d0172140ec734ee | | mBERT + MD w/ Full Fine-Tune | 22.5 | Not recorded | 69d33dd97d0172140ec734ee | | mBERT + MD w/ Adapter Tune | 24.17 | Not recorded | 69d33dd97d0172140ec734ee | | mBERT + MD w/ Prompt Tune | 24.17 | Not recorded | 69d33dd97d0172140ec734ee | | mBERT + MD w/ Prefix Tune | 26.67 | Not recorded | 69d33dd97d0172140ec734ee | ### XLM-R + MD w/ Full Fine-Tune Bias Score (DE): 4.17. Result ID: 69d36b6047f34ce924bb2803. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"XLM-R","Debiasing Strategy":"Multiple-Debias","Tuning":"Full Fine-Tune"}. This record does not assert independent reproduction. ### mBERT + SD Bias Score (DE): 6.67. Result ID: 69d36b6047f34ce924bb27f7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"mBERT","Debiasing Strategy":"Self-Debias"}. This record does not assert independent reproduction. ### XLM-R + MD w/ Prefix Tune Bias Score (DE): 7.5. Result ID: 69d36b6047f34ce924bb2806. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"XLM-R","Debiasing Strategy":"Multiple-Debias","Tuning":"Prefix Tune"}. This record does not assert independent reproduction. ### XLM-R + MD w/ Adapter Tune Bias Score (DE): 9.17. Result ID: 69d36b6047f34ce924bb2804. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"XLM-R","Debiasing Strategy":"Multiple-Debias","Tuning":"Adapter Tune"}. This record does not assert independent reproduction. ### XLM-R + MCDA Bias Score (DE): 11.88. Result ID: 69d36b6047f34ce924bb2801. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"XLM-R","Debiasing Strategy":"Multilingual CDA"}. This record does not assert independent reproduction. ### XLM-R + MD w/ Prompt Tune Bias Score (DE): 13.33. Result ID: 69d36b6047f34ce924bb2805. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"XLM-R","Debiasing Strategy":"Multiple-Debias","Tuning":"Prompt Tune"}. This record does not assert independent reproduction. ### mBERT + MSD Bias Score (DE): 15.83. Result ID: 69d36b6047f34ce924bb27f9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"mBERT","Debiasing Strategy":"Multilingual Self-Debias"}. This record does not assert independent reproduction. ### XLM-R + CDA Bias Score (DE): 16.67. Result ID: 69d36b6047f34ce924bb27ff. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"XLM-R","Debiasing Strategy":"Counterfactual Data Augmentation"}. This record does not assert independent reproduction. ### XLM-R + SD Bias Score (DE): 16.67. Result ID: 69d36b6047f34ce924bb2800. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"XLM-R","Debiasing Strategy":"Self-Debias"}. This record does not assert independent reproduction. ### mBERT + MCDA Bias Score (DE): 17.71. Result ID: 69d36b6047f34ce924bb27f8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"mBERT","Debiasing Strategy":"Multilingual CDA"}. This record does not assert independent reproduction. ### mBERT Bias Score (DE): 19.17. Result ID: 69d36b6047f34ce924bb27f5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Base Model":"mBERT"}. This record does not assert independent reproduction. ### XLM-R + MSD Bias Score (DE): 20. Result ID: 69d36b6047f34ce924bb2802. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"XLM-R","Debiasing Strategy":"Multilingual Self-Debias"}. This record does not assert independent reproduction. ### XLM-R Bias Score (DE): 21.67. Result ID: 69d36b6047f34ce924bb27fe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"XLM-R"}. This record does not assert independent reproduction. ### mBERT + CDA Bias Score (DE): 22.5. Result ID: 69d36b6047f34ce924bb27f6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"mBERT","Debiasing Strategy":"Counterfactual Data Augmentation"}. This record does not assert independent reproduction. ### mBERT + MD w/ Full Fine-Tune Bias Score (DE): 22.5. Result ID: 69d36b6047f34ce924bb27fa. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"mBERT","Debiasing Strategy":"Multiple-Debias","Tuning":"Full Fine-Tune"}. This record does not assert independent reproduction. ### mBERT + MD w/ Adapter Tune Bias Score (DE): 24.17. Result ID: 69d36b6047f34ce924bb27fb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"mBERT","Debiasing Strategy":"Multiple-Debias","Tuning":"Adapter Tune"}. This record does not assert independent reproduction. ### mBERT + MD w/ Prompt Tune Bias Score (DE): 24.17. Result ID: 69d36b6047f34ce924bb27fc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"mBERT","Debiasing Strategy":"Multiple-Debias","Tuning":"Prompt Tune"}. This record does not assert independent reproduction. ### mBERT + MD w/ Prefix Tune Bias Score (DE): 26.67. Result ID: 69d36b6047f34ce924bb27fd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"mBERT","Debiasing Strategy":"Multiple-Debias","Tuning":"Prefix Tune"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.