# Hallucination Evaluation on POPE MSCOCO, A-OKVQA, GQA average (Adversarial) Source: [Hallucination Evaluation on POPE MSCOCO, A-OKVQA, GQA average (Adversarial)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:32:52.674Z Average performance across models evaluated using the POPE methodology (Adversarial setting) on the combined MSCOCO, A-OKVQA, and GQA datasets for hallucination detection, reporting Accuracy, Precision, Recall, and F1 scores. 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) | POPE MSCOCO, A-OKVQA, GQA average (Adversarial) | | Metric | Accuracy | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 15 of 15 | | 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. - [Precision](): Higher is better; unit: PERCENT. - [Recall](): Higher is better; unit: PERCENT. - [F1 Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | Vanilla | 84.11 | Not recorded | 6993f5e69ebc499c974098f1 | | AdaVBoost | 84.09 | Not recorded | 6993f5e69ebc499c974098f1 | | PAI | 83.55 | Not recorded | 6993f5e69ebc499c974098f1 | | VGA | 83.05 | Not recorded | 6993f5e69ebc499c974098f1 | | VAF | 82.39 | Not recorded | 6993f5e69ebc499c974098f1 | | Vanilla | 80.18 | Not recorded | 6993f5e69ebc499c974098f1 | | AdaVBoost | 80.15 | Not recorded | 6993f5e69ebc499c974098f1 | | VGA | 78.71 | Not recorded | 6993f5e69ebc499c974098f1 | | Vanilla | 78.68 | Not recorded | 6993f5e69ebc499c974098f1 | | VAF | 78.48 | Not recorded | 6993f5e69ebc499c974098f1 | | AdaVBoost | 78.46 | Not recorded | 6993f5e69ebc499c974098f1 | | VAF | 77.81 | Not recorded | 6993f5e69ebc499c974098f1 | | VGA | 77.46 | Not recorded | 6993f5e69ebc499c974098f1 | | PAI | 77.23 | Not recorded | 6993f5e69ebc499c974098f1 | | PAI | 77.22 | Not recorded | 6993f5e69ebc499c974098f1 | ### Vanilla Accuracy: 84.11. Result ID: 699a77af5698665ec400e459. [Reporting paper]() Recorded evaluation configuration: {"MLLM":"Qwen3-VL"}. This record does not assert independent reproduction. ### AdaVBoost Accuracy: 84.09. Result ID: 699a77af5698665ec400e45d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"MLLM":"Qwen3-VL"}. This record does not assert independent reproduction. ### PAI Accuracy: 83.55. Result ID: 699a77af5698665ec400e45a. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"MLLM":"Qwen3-VL"}. This record does not assert independent reproduction. ### VGA Accuracy: 83.05. Result ID: 699a77af5698665ec400e45c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"MLLM":"Qwen3-VL"}. This record does not assert independent reproduction. ### VAF Accuracy: 82.39. Result ID: 699a77af5698665ec400e45b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"MLLM":"Qwen3-VL"}. This record does not assert independent reproduction. ### Vanilla Accuracy: 80.18. Result ID: 699a77af5698665ec400e454. [Reporting paper]() Recorded evaluation configuration: {"MLLM":"LLaVA-NeXT"}. This record does not assert independent reproduction. ### AdaVBoost Accuracy: 80.15. Result ID: 699a77af5698665ec400e458. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"MLLM":"LLaVA-NeXT"}. This record does not assert independent reproduction. ### VGA Accuracy: 78.71. Result ID: 699a77af5698665ec400e461. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"MLLM":"InternVL3.5"}. This record does not assert independent reproduction. ### Vanilla Accuracy: 78.68. Result ID: 699a77af5698665ec400e45e. [Reporting paper]() Recorded evaluation configuration: {"MLLM":"InternVL3.5"}. This record does not assert independent reproduction. ### VAF Accuracy: 78.48. Result ID: 699a77af5698665ec400e460. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"MLLM":"InternVL3.5"}. This record does not assert independent reproduction. ### AdaVBoost Accuracy: 78.46. Result ID: 699a77af5698665ec400e462. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"MLLM":"InternVL3.5"}. This record does not assert independent reproduction. ### VAF Accuracy: 77.81. Result ID: 699a77af5698665ec400e456. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"MLLM":"LLaVA-NeXT"}. This record does not assert independent reproduction. ### VGA Accuracy: 77.46. Result ID: 699a77af5698665ec400e457. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"MLLM":"LLaVA-NeXT"}. This record does not assert independent reproduction. ### PAI Accuracy: 77.23. Result ID: 699a77af5698665ec400e455. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"MLLM":"LLaVA-NeXT"}. This record does not assert independent reproduction. ### PAI Accuracy: 77.22. Result ID: 699a77af5698665ec400e45f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"MLLM":"InternVL3.5"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.