# Object Hallucination Detection on POPE (average over three sampling strategies) Source: [Object Hallucination Detection on POPE (average over three sampling strategies)]() Collection index: [llms.txt]() Data updated: 2026-05-13T10:54:39.329Z Evaluation of model performance on detecting object hallucinations using the POPE methodology, specifically reporting results averaged across three distinct sampling strategies, utilizing AUROC and AUPR as primary 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 | Object Hallucination Detection | | Dataset (including any stated split) | POPE average over three sampling strategies | | Metric | AUROC | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 35 of 35 | | 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 - [AUROC](): Higher is better; unit: PERCENT. - [AUPR](): Higher is better; unit: PERCENT. ## Recorded results | Method | AUROC | Result date | Report paper | | --- | --- | --- | --- | | InsLen | 83.94 | Not recorded | 6a0405dfdbc836b87d8278d2 | | InsLen | 79.42 | Not recorded | 6a0405dfdbc836b87d8278d2 | | InsLen | 76.5 | Not recorded | 6a0405dfdbc836b87d8278d2 | | InsLen | 75.42 | Not recorded | 6a0405dfdbc836b87d8278d2 | | InsLen | 74.09 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Entropy | 73.52 | Not recorded | 6a0405dfdbc836b87d8278d2 | | GLSIM | 70.13 | Not recorded | 6a0405dfdbc836b87d8278d2 | | NLL | 68.43 | Not recorded | 6a0405dfdbc836b87d8278d2 | | GLSIM | 68.38 | Not recorded | 6a0405dfdbc836b87d8278d2 | | GLSIM | 68.24 | Not recorded | 6a0405dfdbc836b87d8278d2 | | SVAR | 67.16 | Not recorded | 6a0405dfdbc836b87d8278d2 | | NLL | 64.26 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Internal Conf. | 63.74 | Not recorded | 6a0405dfdbc836b87d8278d2 | | SVAR | 63.62 | Not recorded | 6a0405dfdbc836b87d8278d2 | | SVAR | 61.47 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Contextual Lens | 61.42 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Entropy | 60.73 | Not recorded | 6a0405dfdbc836b87d8278d2 | | GLSIM | 60.6 | Not recorded | 6a0405dfdbc836b87d8278d2 | | NLL | 59.51 | Not recorded | 6a0405dfdbc836b87d8278d2 | | GLSIM | 58.2 | Not recorded | 6a0405dfdbc836b87d8278d2 | | SVAR | 58.07 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Contextual Lens | 56.07 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Internal Conf. | 55.46 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Entropy | 55.35 | Not recorded | 6a0405dfdbc836b87d8278d2 | | NLL | 55.19 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Entropy | 54.65 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Internal Conf. | 54.41 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Contextual Lens | 54.25 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Contextual Lens | 53.46 | Not recorded | 6a0405dfdbc836b87d8278d2 | | NLL | 53.2 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Entropy | 52.59 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Internal Conf. | 52 | Not recorded | 6a0405dfdbc836b87d8278d2 | | SVAR | 51.62 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Contextual Lens | 46.64 | Not recorded | 6a0405dfdbc836b87d8278d2 | | Internal Conf. | 46.39 | Not recorded | 6a0405dfdbc836b87d8278d2 | ### InsLen AUROC: 83.94. Result ID: 6a04580e47f34ce924eaff95. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### InsLen AUROC: 79.42. Result ID: 6a04580e47f34ce924eaff97. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"mPLUG-Owl3-8B"}. This record does not assert independent reproduction. ### InsLen AUROC: 76.5. Result ID: 6a04580e47f34ce924eaff98. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-OneVision1.5-8B"}. This record does not assert independent reproduction. ### InsLen AUROC: 75.42. Result ID: 6a04580e47f34ce924eaff99. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B"}. This record does not assert independent reproduction. ### InsLen AUROC: 74.09. Result ID: 6a04580e47f34ce924eaff96. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"InstructBLIP-7B"}. This record does not assert independent reproduction. ### Entropy AUROC: 73.52. Result ID: 6a04580e47f34ce924eaff80. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B"}. This record does not assert independent reproduction. ### GLSIM AUROC: 70.13. Result ID: 6a04580e47f34ce924eaff90. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### NLL AUROC: 68.43. Result ID: 6a04580e47f34ce924eaff7a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-OneVision1.5-8B"}. This record does not assert independent reproduction. ### GLSIM AUROC: 68.38. Result ID: 6a04580e47f34ce924eaff92. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"mPLUG-Owl3-8B"}. This record does not assert independent reproduction. ### GLSIM AUROC: 68.24. Result ID: 6a04580e47f34ce924eaff94. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B"}. This record does not assert independent reproduction. ### SVAR AUROC: 67.16. Result ID: 6a04580e47f34ce924eaff88. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"mPLUG-Owl3-8B"}. This record does not assert independent reproduction. ### NLL AUROC: 64.26. Result ID: 6a04580e47f34ce924eaff77. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### Internal Conf. AUROC: 63.74. Result ID: 6a04580e47f34ce924eaff84. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-OneVision1.5-8B"}. This record does not assert independent reproduction. ### SVAR AUROC: 63.62. Result ID: 6a04580e47f34ce924eaff89. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-OneVision1.5-8B"}. This record does not assert independent reproduction. ### SVAR AUROC: 61.47. Result ID: 6a04580e47f34ce924eaff8a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B"}. This record does not assert independent reproduction. ### Contextual Lens AUROC: 61.42. Result ID: 6a04580e47f34ce924eaff8f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B"}. This record does not assert independent reproduction. ### Entropy AUROC: 60.73. Result ID: 6a04580e47f34ce924eaff7f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-OneVision1.5-8B"}. This record does not assert independent reproduction. ### GLSIM AUROC: 60.6. Result ID: 6a04580e47f34ce924eaff93. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-OneVision1.5-8B"}. This record does not assert independent reproduction. ### NLL AUROC: 59.51. Result ID: 6a04580e47f34ce924eaff79. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"mPLUG-Owl3-8B"}. This record does not assert independent reproduction. ### GLSIM AUROC: 58.2. Result ID: 6a04580e47f34ce924eaff91. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"InstructBLIP-7B"}. This record does not assert independent reproduction. ### SVAR AUROC: 58.07. Result ID: 6a04580e47f34ce924eaff87. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"InstructBLIP-7B"}. This record does not assert independent reproduction. ### Contextual Lens AUROC: 56.07. Result ID: 6a04580e47f34ce924eaff8c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"InstructBLIP-7B"}. This record does not assert independent reproduction. ### Internal Conf. AUROC: 55.46. Result ID: 6a04580e47f34ce924eaff83. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"mPLUG-Owl3-8B"}. This record does not assert independent reproduction. ### Entropy AUROC: 55.35. Result ID: 6a04580e47f34ce924eaff7e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"mPLUG-Owl3-8B"}. This record does not assert independent reproduction. ### NLL AUROC: 55.19. Result ID: 6a04580e47f34ce924eaff7b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B"}. This record does not assert independent reproduction. ### Entropy AUROC: 54.65. Result ID: 6a04580e47f34ce924eaff7d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"InstructBLIP-7B"}. This record does not assert independent reproduction. ### Internal Conf. AUROC: 54.41. Result ID: 6a04580e47f34ce924eaff81. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### Contextual Lens AUROC: 54.25. Result ID: 6a04580e47f34ce924eaff8e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-OneVision1.5-8B"}. This record does not assert independent reproduction. ### Contextual Lens AUROC: 53.46. Result ID: 6a04580e47f34ce924eaff8b. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### NLL AUROC: 53.2. Result ID: 6a04580e47f34ce924eaff78. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"InstructBLIP-7B"}. This record does not assert independent reproduction. ### Entropy AUROC: 52.59. Result ID: 6a04580e47f34ce924eaff7c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### Internal Conf. AUROC: 52. Result ID: 6a04580e47f34ce924eaff82. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"InstructBLIP-7B"}. This record does not assert independent reproduction. ### SVAR AUROC: 51.62. Result ID: 6a04580e47f34ce924eaff86. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"LLaVA-1.5-7B"}. This record does not assert independent reproduction. ### Contextual Lens AUROC: 46.64. Result ID: 6a04580e47f34ce924eaff8d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"mPLUG-Owl3-8B"}. This record does not assert independent reproduction. ### Internal Conf. AUROC: 46.39. Result ID: 6a04580e47f34ce924eaff85. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen3-VL-8B"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.