# Semantic Verifier Evaluation on SceneCritic Source: [Semantic Verifier Evaluation on SceneCritic]() Collection index: [llms.txt]() Data updated: 2026-04-15T10:22:56.404Z Evaluation of Semantic Verifier performance on the SceneCritic dataset. 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 | Semantic Verifier Evaluation | | Dataset (including any stated split) | SceneCritic | | Metric | Semantic Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 41 of 41 | | 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 - [Semantic Score](): Higher is better; unit: PERCENT. - [Original Metric Score](): Higher is better; unit: PERCENT. - [Proximal Overlap](): Higher is better; unit: PERCENT. - [True Overlap](): Higher is better; unit: PERCENT. - [Average Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Semantic Score | Result date | Report paper | | --- | --- | --- | --- | | DeepSeek-3.2V | 82.6 | Not recorded | 69df1bcb7d0172140ec7bc2c | | DeepSeek-3.2V | 78.9 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Gemini-2.5-flash | 78.7 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen3-235B | 76.9 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Gemini-2.5-flash | 76.7 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Gemini-2.5-flash | 76.3 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Gemini-2.5-flash | 75.8 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen3-235B | 75 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Gemini-2.5-flash | 74.2 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen3-14B | 73.4 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen3-14B | 73.3 | Not recorded | 69df1bcb7d0172140ec7bc2c | | VL-Reasoner-72B | 72.8 | Not recorded | 69df1bcb7d0172140ec7bc2c | | VL-Reasoner-72B | 72.8 | Not recorded | 69df1bcb7d0172140ec7bc2c | | LLaMA4 Maverick | 72.6 | Not recorded | 69df1bcb7d0172140ec7bc2c | | UI-Venus-Navi-72B | 72.6 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen2.5-72B-VL | 72.5 | Not recorded | 69df1bcb7d0172140ec7bc2c | | VL-Reasoner-72B | 72.4 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen2.5-72B-VL | 72.3 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen2.5-72B-VL | 72.2 | Not recorded | 69df1bcb7d0172140ec7bc2c | | UI-Venus-Navi-72B | 72.2 | Not recorded | 69df1bcb7d0172140ec7bc2c | | UI-Venus-Navi-72B | 72 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen2.5-72B-VL | 71.2 | Not recorded | 69df1bcb7d0172140ec7bc2c | | VL-Reasoner-72B | 71 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen2.5-72B-VL | 71 | Not recorded | 69df1bcb7d0172140ec7bc2c | | UI-Venus-Navi-72B | 70.7 | Not recorded | 69df1bcb7d0172140ec7bc2c | | LLaMA4 Maverick | 69.9 | Not recorded | 69df1bcb7d0172140ec7bc2c | | VL-Reasoner-72B | 68.9 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Holo1.5-72B | 68.5 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen3-14B-Intuitor-MATH-1EPOCH | 68.4 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Holo1.5-72B | 68.3 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen2.5-14B-GRPO | 67.9 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Holo1.5-72B | 67.9 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Holo1.5-72B | 67.6 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen2.5-14B-GRPO | 67.3 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Holo1.5-72B | 66.9 | Not recorded | 69df1bcb7d0172140ec7bc2c | | UI-Venus-Navi-72B | 66.5 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen3-14B-Intuitor-MATH-1EPOCH | 65.8 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen2.5-VL-7B-MM-RLHF | 65.5 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen2.5-VL-7B-MM-RLHF | 64 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen3-14B-GRPO-MATH-1EPOCH | 63.8 | Not recorded | 69df1bcb7d0172140ec7bc2c | | Qwen3-14B-GRPO-MATH-1EPOCH | 63.8 | Not recorded | 69df1bcb7d0172140ec7bc2c | ### DeepSeek-3.2V Semantic Score: 82.6. Result ID: 69df667247f34ce924c8bec0. [Reporting paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### DeepSeek-3.2V Semantic Score: 78.9. Result ID: 69df667247f34ce924c8becd. [Reporting paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### Gemini-2.5-flash Semantic Score: 78.7. Result ID: 69df667247f34ce924c8bed8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Image"}. This record does not assert independent reproduction. ### Qwen3-235B Semantic Score: 76.9. Result ID: 69df667247f34ce924c8becf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### Gemini-2.5-flash Semantic Score: 76.7. Result ID: 69df667247f34ce924c8bedd. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Img+Text"}. This record does not assert independent reproduction. ### Gemini-2.5-flash Semantic Score: 76.3. Result ID: 69df667247f34ce924c8bebe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Gemini-2.5-flash Semantic Score: 75.8. Result ID: 69df667247f34ce924c8bee2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Sem+Text"}. This record does not assert independent reproduction. ### Qwen3-235B Semantic Score: 75. Result ID: 69df667247f34ce924c8bec2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Gemini-2.5-flash Semantic Score: 74.2. Result ID: 69df667247f34ce924c8becb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### Qwen3-14B Semantic Score: 73.4. Result ID: 69df667247f34ce924c8bec3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Qwen3-14B Semantic Score: 73.3. Result ID: 69df667247f34ce924c8bed0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### VL-Reasoner-72B Semantic Score: 72.8. Result ID: 69df667247f34ce924c8bed6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### VL-Reasoner-72B Semantic Score: 72.8. Result ID: 69df667247f34ce924c8bee5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Sem+Text"}. This record does not assert independent reproduction. ### LLaMA4 Maverick Semantic Score: 72.6. Result ID: 69df667247f34ce924c8bec1. [Reporting paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### UI-Venus-Navi-72B Semantic Score: 72.6. Result ID: 69df667247f34ce924c8bed7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### Qwen2.5-72B-VL Semantic Score: 72.5. Result ID: 69df667247f34ce924c8becc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### VL-Reasoner-72B Semantic Score: 72.4. Result ID: 69df667247f34ce924c8bee0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Img+Text"}. This record does not assert independent reproduction. ### Qwen2.5-72B-VL Semantic Score: 72.3. Result ID: 69df667247f34ce924c8bee3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Sem+Text"}. This record does not assert independent reproduction. ### Qwen2.5-72B-VL Semantic Score: 72.2. Result ID: 69df667247f34ce924c8bebf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### UI-Venus-Navi-72B Semantic Score: 72.2. Result ID: 69df667247f34ce924c8bedc. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Image"}. This record does not assert independent reproduction. ### UI-Venus-Navi-72B Semantic Score: 72. Result ID: 69df667247f34ce924c8bee1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Img+Text"}. This record does not assert independent reproduction. ### Qwen2.5-72B-VL Semantic Score: 71.2. Result ID: 69df667247f34ce924c8bed9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Image"}. This record does not assert independent reproduction. ### VL-Reasoner-72B Semantic Score: 71. Result ID: 69df667247f34ce924c8bedb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Image"}. This record does not assert independent reproduction. ### Qwen2.5-72B-VL Semantic Score: 71. Result ID: 69df667247f34ce924c8bede. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Img+Text"}. This record does not assert independent reproduction. ### UI-Venus-Navi-72B Semantic Score: 70.7. Result ID: 69df667247f34ce924c8bee6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Sem+Text"}. This record does not assert independent reproduction. ### LLaMA4 Maverick Semantic Score: 69.9. Result ID: 69df667247f34ce924c8bece. [Reporting paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### VL-Reasoner-72B Semantic Score: 68.9. Result ID: 69df667247f34ce924c8bec9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Holo1.5-72B Semantic Score: 68.5. Result ID: 69df667247f34ce924c8beda. [Reporting paper]() Recorded evaluation configuration: {"Refinement Variant":"Image"}. This record does not assert independent reproduction. ### Qwen3-14B-Intuitor-MATH-1EPOCH Semantic Score: 68.4. Result ID: 69df667247f34ce924c8bed4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### Holo1.5-72B Semantic Score: 68.3. Result ID: 69df667247f34ce924c8bec8. [Reporting paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Qwen2.5-14B-GRPO Semantic Score: 67.9. Result ID: 69df667247f34ce924c8bed2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### Holo1.5-72B Semantic Score: 67.9. Result ID: 69df667247f34ce924c8bedf. [Reporting paper]() Recorded evaluation configuration: {"Refinement Variant":"Img+Text"}. This record does not assert independent reproduction. ### Holo1.5-72B Semantic Score: 67.6. Result ID: 69df667247f34ce924c8bee4. [Reporting paper]() Recorded evaluation configuration: {"Refinement Variant":"Sem+Text"}. This record does not assert independent reproduction. ### Qwen2.5-14B-GRPO Semantic Score: 67.3. Result ID: 69df667247f34ce924c8bec5. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Holo1.5-72B Semantic Score: 66.9. Result ID: 69df667247f34ce924c8bed5. [Reporting paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### UI-Venus-Navi-72B Semantic Score: 66.5. Result ID: 69df667247f34ce924c8beca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Qwen3-14B-Intuitor-MATH-1EPOCH Semantic Score: 65.8. Result ID: 69df667247f34ce924c8bec7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B-MM-RLHF Semantic Score: 65.5. Result ID: 69df667247f34ce924c8bed1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B-MM-RLHF Semantic Score: 64. Result ID: 69df667247f34ce924c8bec4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Qwen3-14B-GRPO-MATH-1EPOCH Semantic Score: 63.8. Result ID: 69df667247f34ce924c8bec6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"Heuristic"}. This record does not assert independent reproduction. ### Qwen3-14B-GRPO-MATH-1EPOCH Semantic Score: 63.8. Result ID: 69df667247f34ce924c8bed3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Refinement Variant":"LLM"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.