# Viewpoint-Aware Consistency Reasoning on COMFORT Multi Source: [Viewpoint-Aware Consistency Reasoning on COMFORT Multi]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:48:34.818Z Evaluation of model performance on the COMFORT Multi dataset, focusing on Viewpoint-Aware Consistency Reasoning capabilities, measured across viewpoint consistency 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 | Viewpoint-Aware Consistency Reasoning | | Dataset (including any stated split) | COMFORT Multi | | Metric | Left/Right Consistency | | 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 - [Left/Right Consistency](): Higher is better; unit: PERCENT. - [Closer Consistency](): Higher is better; unit: PERCENT. - [Visibility Consistency](): Higher is better; unit: PERCENT. - [Facing Consistency](): Higher is better; unit: PERCENT. ## Recorded results | Method | Left/Right Consistency | Result date | Report paper | | --- | --- | --- | --- | | SymPL | 76 | Not recorded | 699d30776017c4460ffab35c | | Qwen2.5-VL | 72.5 | Not recorded | 699d30776017c4460ffab35c | | Qwen2.5-VL | 67.5 | Not recorded | 699d30776017c4460ffab35c | | Qwen2.5-VL | 66.5 | Not recorded | 699d30776017c4460ffab35c | | Qwen2.5-VL | 57.5 | Not recorded | 699d30776017c4460ffab35c | | APC-Vis | 53 | Not recorded | 699d30776017c4460ffab35c | | APC-Num | 44.5 | Not recorded | 699d30776017c4460ffab35c | ### SymPL Left/Right Consistency: 76. Result ID: 699e38075698665ec4f470ec. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-VL Left/Right Consistency: 72.5. Result ID: 699e38075698665ec4f470e8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"prompting":"SoM"}. This record does not assert independent reproduction. ### Qwen2.5-VL Left/Right Consistency: 67.5. Result ID: 699e38075698665ec4f470e6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen2.5-VL Left/Right Consistency: 66.5. Result ID: 699e38075698665ec4f470e9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"prompting":"SCAFFOLD"}. This record does not assert independent reproduction. ### Qwen2.5-VL Left/Right Consistency: 57.5. Result ID: 699e38075698665ec4f470e7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"prompting":"CoT"}. This record does not assert independent reproduction. ### APC-Vis Left/Right Consistency: 53. Result ID: 699e38075698665ec4f470eb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### APC-Num Left/Right Consistency: 44.5. Result ID: 699e38075698665ec4f470ea. [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.