# Text-to-image preference evaluation on HPD v3 (test) Source: [Text-to-image preference evaluation on HPD v3 (test)]() Collection index: [llms.txt]() Data updated: 2026-05-12T08:24:17.507Z Evaluation of Text-to-Image model outputs based on human preference accuracy derived from the HPD v3 test set. 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 | Text-to-image preference evaluation | | Dataset (including any stated split) | HPD v3 (test) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 11 of 11 | | 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 - [Accuracy](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Gemini 3.1 Pro | 78.3 | Not recorded | 6a02b424dbc836b87d8246e4 | | HPSv3 | 76.9 | Not recorded | 6a02b424dbc836b87d8246e4 | | Gemini 3.1 Pro | 76.6 | Not recorded | 6a02b424dbc836b87d8246e4 | | GPT-5 | 76.1 | Not recorded | 6a02b424dbc836b87d8246e4 | | GPT-5 | 72.4 | Not recorded | 6a02b424dbc836b87d8246e4 | | Qwen3-VL-8B | 70.2 | Not recorded | 6a02b424dbc836b87d8246e4 | | UnifiedReward-Thinking | 68.1 | Not recorded | 6a02b424dbc836b87d8246e4 | | Qwen3-VL-8B | 67.2 | Not recorded | 6a02b424dbc836b87d8246e4 | | UnifiedReward | 66 | Not recorded | 6a02b424dbc836b87d8246e4 | | PickScore | 65.6 | Not recorded | 6a02b424dbc836b87d8246e4 | | ImageReward | 58.6 | Not recorded | 6a02b424dbc836b87d8246e4 | ### Gemini 3.1 Pro Accuracy: 78.3. Result ID: 6a02e39f47f34ce924e59eb7. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"ARR":"true"}. This record does not assert independent reproduction. ### HPSv3 Accuracy: 76.9. Result ID: 6a02e39f47f34ce924e59eb1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Gemini 3.1 Pro Accuracy: 76.6. Result ID: 6a02e39f47f34ce924e59eb6. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"ARR":"false"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 76.1. Result ID: 6a02e39f47f34ce924e59eb5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"ARR":"true"}. This record does not assert independent reproduction. ### GPT-5 Accuracy: 72.4. Result ID: 6a02e39f47f34ce924e59eb4. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"ARR":"false"}. This record does not assert independent reproduction. ### Qwen3-VL-8B Accuracy: 70.2. Result ID: 6a02e39f47f34ce924e59eb3. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"ARR":"true"}. This record does not assert independent reproduction. ### UnifiedReward-Thinking Accuracy: 68.1. Result ID: 6a02e39f47f34ce924e59eb0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### Qwen3-VL-8B Accuracy: 67.2. Result ID: 6a02e39f47f34ce924e59eb2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"ARR":"false"}. This record does not assert independent reproduction. ### UnifiedReward Accuracy: 66. Result ID: 6a02e39f47f34ce924e59eaf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### PickScore Accuracy: 65.6. Result ID: 6a02e39f47f34ce924e59ead. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: Not recorded. This record does not assert independent reproduction. ### ImageReward Accuracy: 58.6. Result ID: 6a02e39f47f34ce924e59eae. [Reporting paper]() [Method origin paper]() [Implementation]() 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.