# Hallucination Evaluation on HaloQuest Source: [Hallucination Evaluation on HaloQuest]() Collection index: [llms.txt]() Data updated: 2026-03-19T09:47:59.602Z Evaluation of model performance on hallucination evaluation using the HaloQuest dataset, measured by ACC%, S, and I. 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 | Hallucination Evaluation | | Dataset (including any stated split) | HaloQuest | | Metric | Score S | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 22 of 22 | | 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 - [Score S](): Higher is better; unit: PERCENT. - [Accuracy](): Higher is better; unit: PERCENT. - [Score I](): Higher is better; unit: PERCENT. ## Recorded results | Method | Score S | Result date | Report paper | | --- | --- | --- | --- | | Qwen2.5-VL + FINER-Tuning | 80.8 | Not recorded | 69bb83290636e5bf4afc1c13 | | Qwen2.5-VL | 74.8 | Not recorded | 69bb83290636e5bf4afc1c13 | | InternVL-3.5 + FINER-Tuning | 73.5 | Not recorded | 69bb83290636e5bf4afc1c13 | | InternVL-3.5 + FINER-Tuning | 71 | Not recorded | 69bb83290636e5bf4afc1c13 | | InternVL-3.5 | 65 | Not recorded | 69bb83290636e5bf4afc1c13 | | LLaVA-1.6 + FINER-Tuning | 63.5 | Not recorded | 69bb83290636e5bf4afc1c13 | | OmniLMM + RLAIF-V | 62.4 | Not recorded | 69bb83290636e5bf4afc1c13 | | InternVL-3.5 | 62.4 | Not recorded | 69bb83290636e5bf4afc1c13 | | Qwen2.5-VL-7B+VRGA | 54.9 | Not recorded | 69b8e0390636e5bf4afbed13 | | Qwen2.5-VL-7B | 50.2 | Not recorded | 69b8e0390636e5bf4afbed13 | | Qwen2.5-VL-3B+VRGA | 48.8 | Not recorded | 69b8e0390636e5bf4afbed13 | | Qwen3-VL-30B+VRGA | 46.5 | Not recorded | 69b8e0390636e5bf4afbed13 | | Qwen3-VL-30B | 45.4 | Not recorded | 69b8e0390636e5bf4afbed13 | | Qwen2.5-VL-3B | 44.5 | Not recorded | 69b8e0390636e5bf4afbed13 | | LLaVA-1.6 | 44.2 | Not recorded | 69bb83290636e5bf4afc1c13 | | Qwen2-VL-7B | 40.6 | Not recorded | 69b8e0390636e5bf4afbed13 | | Qwen2-VL-7B+VRGA | 40.4 | Not recorded | 69b8e0390636e5bf4afbed13 | | OmniLMM | 39.9 | Not recorded | 69bb83290636e5bf4afc1c13 | | Qwen2-VL-7B+ICOT | 36.7 | Not recorded | 69b8e0390636e5bf4afbed13 | ### Qwen2.5-VL + FINER-Tuning Score S: 80.8. Result ID: 69bbc64947f34ce924a61b87. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Size":"7B"}. This record does not assert independent reproduction. ### Qwen2.5-VL Score S: 74.8. Result ID: 69bbc64947f34ce924a61b86. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"7B"}. This record does not assert independent reproduction. ### InternVL-3.5 + FINER-Tuning Score S: 73.5. Result ID: 69bbc64947f34ce924a61b89. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Size":"8B"}. This record does not assert independent reproduction. ### InternVL-3.5 + FINER-Tuning Score S: 71. Result ID: 69bbc64947f34ce924a61b8b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Size":"14B"}. This record does not assert independent reproduction. ### InternVL-3.5 Score S: 65. Result ID: 69bbc64947f34ce924a61b8a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"14B"}. This record does not assert independent reproduction. ### LLaVA-1.6 + FINER-Tuning Score S: 63.5. Result ID: 69bbc64947f34ce924a61b85. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Size":"7B"}. This record does not assert independent reproduction. ### OmniLMM + RLAIF-V Score S: 62.4. Result ID: 69bbc64947f34ce924a61b83. [Reporting paper]() Recorded evaluation configuration: {"Size":"12B"}. This record does not assert independent reproduction. ### InternVL-3.5 Score S: 62.4. Result ID: 69bbc64947f34ce924a61b88. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Size":"8B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B+VRGA Score S: 54.9. Result ID: 69b9232147f34ce924a238a1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B","Enhancement":"VRGA"}. This record does not assert independent reproduction. ### Qwen2.5-VL-7B Score S: 50.2. Result ID: 69b9232147f34ce924a2389f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-7B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-3B+VRGA Score S: 48.8. Result ID: 69b9232147f34ce924a2389e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B","Enhancement":"VRGA"}. This record does not assert independent reproduction. ### Qwen3-VL-30B+VRGA Score S: 46.5. Result ID: 69b9232147f34ce924a2389b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL-30B","Enhancement":"VRGA"}. This record does not assert independent reproduction. ### Qwen3-VL-30B Score S: 45.4. Result ID: 69b9232147f34ce924a2389a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen3-VL-30B"}. This record does not assert independent reproduction. ### Qwen2.5-VL-3B Score S: 44.5. Result ID: 69b9232147f34ce924a2389c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2.5-VL-3B"}. This record does not assert independent reproduction. ### LLaVA-1.6 Score S: 44.2. Result ID: 69bbc64947f34ce924a61b84. [Reporting paper]() Recorded evaluation configuration: {"Size":"7B"}. This record does not assert independent reproduction. ### Qwen2-VL-7B Score S: 40.6. Result ID: 69b9232147f34ce924a238a2. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Backbone":"Qwen2-VL-7B"}. This record does not assert independent reproduction. ### Qwen2-VL-7B+VRGA Score S: 40.4. Result ID: 69b9232147f34ce924a238a5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2-VL-7B","Enhancement":"VRGA"}. This record does not assert independent reproduction. ### OmniLMM Score S: 39.9. Result ID: 69bbc64947f34ce924a61b82. [Reporting paper]() Recorded evaluation configuration: {"Size":"12B"}. This record does not assert independent reproduction. ### Qwen2-VL-7B+ICOT Score S: 36.7. Result ID: 69b9232147f34ce924a238a4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Backbone":"Qwen2-VL-7B","Enhancement":"ICOT"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.