# Suitability violation and drift analysis on Extended cross-model comparison evaluation dataset Source: [Suitability violation and drift analysis on Extended cross-model comparison evaluation dataset]() Collection index: [llms.txt]() Data updated: 2026-03-16T09:41:57.026Z Evaluation of model suitability violation and drift analysis performance using the Extended cross-model comparison evaluation dataset, measured by metrics such as SVRs, D-bar, and AR. 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 | Suitability violation and drift analysis | | Dataset (including any stated split) | Extended cross-model comparison evaluation dataset | | Metric | SVRs | | Unit | PERCENT | | 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 - [SVRs](): Higher is better; unit: PERCENT. - [SVRr](): Higher is better; unit: PERCENT. - [Severity SVR](): Higher is better; unit: SCALAR. - [D-bar](): Lower is better; unit: PERCENT. - [AR](): Higher is better; unit: SCALAR. - [MDR](): Lower is better; unit: PERCENT. - [UPR](): Higher is better; unit: SCALAR. ## Recorded results | Method | SVRs | Result date | Report paper | | --- | --- | --- | --- | | Claude Sonnet 4.6 | 92.6 | Not recorded | 69b78e560636e5bf4afbd9e8 | | Mistral Large 3 | 90 | Not recorded | 69b78e560636e5bf4afbd9e8 | | GPT-5.2 | 88.3 | Not recorded | 69b78e560636e5bf4afbd9e8 | | Gemma 3 12B-IT | 87.4 | Not recorded | 69b78e560636e5bf4afbd9e8 | | Qwen3-32B | 80 | Not recorded | 69b78e560636e5bf4afbd9e8 | | Ministral 3 14B | 67.8 | Not recorded | 69b78e560636e5bf4afbd9e8 | | Qwen2.5-7B | 64.8 | Not recorded | 69b78e560636e5bf4afbd9e8 | ### Claude Sonnet 4.6 SVRs: 92.6. Result ID: 69b7d06247f34ce9249fe374. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Mistral Large 3 SVRs: 90. Result ID: 69b7d06247f34ce9249fe376. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### GPT-5.2 SVRs: 88.3. Result ID: 69b7d06247f34ce9249fe373. [Reporting paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Gemma 3 12B-IT SVRs: 87.4. Result ID: 69b7d06247f34ce9249fe372. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen3-32B SVRs: 80. Result ID: 69b7d06247f34ce9249fe370. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Ministral 3 14B SVRs: 67.8. Result ID: 69b7d06247f34ce9249fe375. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-7B SVRs: 64.8. Result ID: 69b7d06247f34ce9249fe371. [Reporting paper]() [Method origin 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.