Object Hallucination Detection on MSCOCO
89.62AUROCInsLen
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| InsLenModel=LLaVA-1.5-13B2026.05 | 89.62 | — | — | — | — | — | 96.27 | |
| InsLenModel=InstructBLIP-13B2026.05 | 89.04 | — | — | — | — | — | 95.46 | |
| InsLenModel=LLaVA-1.5-7B2026.05 | 86.93 | — | — | — | — | — | 96.38 | |
| InsLenModel=InstructBLIP-7B2026.05 | 85.5 | — | — | — | — | — | 94.74 | |
| GLSIMModel=LLaVA-1.5-13B2026.05 | 85.36 | — | — | — | — | — | 95.37 | |
| GLSIMModel=InstructBLIP-13B2026.05 | 84.77 | — | — | — | — | — | 95.01 | |
| GLSIMModel=LLaVA-1.5-7B2026.05 | 83.24 | — | — | — | — | — | 93.81 | |
| InsLenModel=LLaVA-OneVision1.5-8B2026.05 | 82.41 | — | — | — | — | — | 96.99 | |
| GLSIMModel=InstructBLIP-7B2026.05 | 82.38 | — | — | — | — | — | 93.57 | |
| Internal Conf.Model=InstructBLIP-7B2026.05 | 81.62 | — | — | — | — | — | 93.27 | |
| InsLenModel=Qwen3-VL-8B2026.05 | 81.02 | — | — | — | — | — | 96.7 | |
| NLLModel=InstructBLIP-7B2026.05 | 78.41 | — | — | — | — | — | 90.57 | |
| SVARModel=InstructBLIP-7B2026.05 | 77.59 | — | — | — | — | — | 92.63 | |
| SVARModel=Qwen3-VL-8B2026.05 | 75.36 | — | — | — | — | — | 93.44 | |
| InsLenModel=mPLUG-Owl3-8B2026.05 | 75.11 | — | — | — | — | — | 96.47 | |
| SVARModel=LLaVA-1.5-7B2026.05 | 73.97 | — | — | — | — | — | 91.33 | |
| Contextual LensModel=InstructBLIP-7B2026.05 | 73.91 | — | — | — | — | — | 93.04 | |
| Internal Conf.Model=LLaVA-1.5-13B2026.05 | 73.66 | — | — | — | — | — | 91.67 | |
| Internal Conf.Model=LLaVA-1.5-7B2026.05 | 73.45 | — | — | — | — | — | 91.21 | |
| GLSIMModel=Qwen3-VL-8B2026.05 | 73.32 | — | — | — | — | — | 93.41 | |
| SVARModel=LLaVA-OneVision1.5-8B2026.05 | 73.08 | — | — | — | — | — | 95.13 | |
| Internal Conf.Model=InstructBLIP-13B2026.05 | 72.34 | — | — | — | — | — | 89.9 | |
| Internal Conf.Model=LLaVA-OneVision1.5-8B2026.05 | 71.89 | — | — | — | — | — | 93.07 | |
| EASYModel=Qwen3-VL-8B2026.05 | 71.02 | — | — | — | — | — | 92.89 | |
| Contextual LensModel=LLaVA-1.5-7B2026.05 | 70.83 | — | — | — | — | — | 87.88 | |
| EASYModel=LLaVA-OneVision1.5-8B2026.05 | 70.43 | — | — | — | — | — | 93.37 | |
| SVARModel=mPLUG-Owl3-8B2026.05 | 70.36 | — | — | — | — | — | 95.66 | |
| EASYModel=LLaVA-1.5-7B2026.05 | 68.05 | — | — | — | — | — | 85.29 | |
| GLSIMModel=LLaVA-OneVision1.5-8B2026.05 | 68.03 | — | — | — | — | — | 93.24 | |
| EntropyModel=InstructBLIP-7B2026.05 | 67.78 | — | — | — | — | — | 80.73 | |
| Internal Conf.Model=Qwen3-VL-8B2026.05 | 67.68 | — | — | — | — | — | 92.03 | |
| GLSIMModel=mPLUG-Owl3-8B2026.05 | 67.12 | — | — | — | — | — | 94.31 | |
| EASYModel=mPLUG-Owl3-8B2026.05 | 66.3 | — | — | — | — | — | 92.43 | |
| EntropyModel=Qwen3-VL-8B2026.05 | 66.23 | — | — | — | — | — | 91.89 | |
| NLLModel=mPLUG-Owl3-8B2026.05 | 66.14 | — | — | — | — | — | 93.27 | |
| EASYModel=InstructBLIP-7B2026.05 | 65.81 | — | — | — | — | — | 79 | |
| EntropyModel=LLaVA-OneVision1.5-8B2026.05 | 65.56 | — | — | — | — | — | 86.79 | |
| NLLModel=LLaVA-1.5-7B2026.05 | 65.19 | — | — | — | — | — | 84.15 | |
| Internal Conf.Model=mPLUG-Owl3-8B2026.05 | 65.07 | — | — | — | — | — | 94.4 | |
| NLLModel=LLaVA-OneVision1.5-8B2026.05 | 64.33 | — | — | — | — | — | 87.2 | |
| NLLModel=Qwen3-VL-8B2026.05 | 64.12 | — | — | — | — | — | 91.49 | |
| Contextual LensModel=Qwen3-VL-8B2026.05 | 63.35 | — | — | — | — | — | 88.43 | |
| EntropyModel=LLaVA-1.5-7B2026.05 | 63.29 | — | — | — | — | — | 82.21 | |
| Contextual LensModel=LLaVA-OneVision1.5-8B2026.05 | 61.02 | — | — | — | — | — | 92.61 | |
| Contextual LensModel=mPLUG-Owl3-8B2026.05 | 57.55 | — | — | — | — | — | 91.61 | |
| EntropyModel=mPLUG-Owl3-8B2026.05 | 57.32 | — | — | — | — | — | 88.83 | |
| EFUFBase Model=LLaVA-v1.5-7B, Setting=Origin2026.01 | — | 41.8 | 18.3 | 24 | 7.7 | 85.9 | — | |
| EFUFBase Model=LLaVA-v1.5-7B, Setting=w/ Relearn 602026.01 | — | 45.8 | 19.7 | 26 | 10.4 | 85.3 | — | |
| EFUFBase Model=LLaVA-v1.5-7B, Setting=w/ Relearn 1002026.01 | — | 46.5 | 19.6 | 28 | 11 | 84.3 | — | |
| EFUFBase Model=LLaVA-v1.5-7B, Setting=w/ Relearn 1402026.01 | — | 48.9 | 20.9 | 29 | 11.8 | 84 | — | |
| EFUFBase Model=LLaVA-v1.5-7B, Setting=w/ LoRA FT2026.01 | — | 48.2 | 20.8 | 27 | 10.8 | 86.1 | — | |
| EFUFBase Model=LLaVA-v1.5-7B, Setting=w/ Adversarial2026.01 | — | 50.5 | 23.9 | 28 | 11 | 68.5 | — | |
| EFUFBase Model=mPLUG-Owl-7B, Setting=Origin2026.01 | — | 43.6 | 24.2 | 46 | 17.7 | 83.8 | — | |
| EFUFBase Model=mPLUG-Owl-7B, Setting=w/ Relearn 602026.01 | — | 50.5 | 26.3 | 50 | 18.5 | 84.8 | — | |
| EFUFBase Model=mPLUG-Owl-7B, Setting=w/ Relearn 1002026.01 | — | 55.2 | 28.6 | 52 | 18.9 | 82.1 | — | |
| EFUFBase Model=mPLUG-Owl-7B, Setting=w/ Relearn 1402026.01 | — | 59.1 | 30.4 | 53 | 20 | 81.4 | — | |
| EFUFBase Model=mPLUG-Owl-7B, Setting=w/ LoRA FT2026.01 | — | 58.6 | 29.7 | 52 | 20.3 | 82 | — | |
| EFUFBase Model=mPLUG-Owl-7B, Setting=w/ Adversarial2026.01 | — | 50 | 26.9 | 48 | 18.5 | 74 | — | |
| SAREBase Model=LLaVA-v1.5-7B, Setting=Origin2026.01 | — | 34.4 | 14.7 | 21 | 6.5 | 86.7 | — | |
| SAREBase Model=LLaVA-v1.5-7B, Setting=w/ Relearn 602026.01 | — | 34.6 | 14.9 | 21 | 6.7 | 86.3 | — | |
| SAREBase Model=LLaVA-v1.5-7B, Setting=w/ Relearn 1002026.01 | — | 35.4 | 15.3 | 22 | 7.9 | 86.2 | — | |
| SAREBase Model=LLaVA-v1.5-7B, Setting=w/ Relearn 1402026.01 | — | 34.6 | 14.8 | 21 | 7.4 | 86.1 | — | |
| SAREBase Model=LLaVA-v1.5-7B, Setting=w/ LoRA FT2026.01 | — | 37.9 | 17.4 | 24 | 9.5 | 86.4 | — | |
| SAREBase Model=LLaVA-v1.5-7B, Setting=w/ Adversarial2026.01 | — | 39.9 | 17.8 | 24 | 9.8 | 67.7 | — | |
| SAREBase Model=mPLUG-Owl-7B, Setting=Origin2026.01 | — | 37.3 | 20.8 | 38 | 14.2 | 82.3 | — | |
| SAREBase Model=mPLUG-Owl-7B, Setting=w/ Relearn 602026.01 | — | 42.3 | 21.8 | 40 | 16.3 | 84.8 | — | |
| SAREBase Model=mPLUG-Owl-7B, Setting=w/ Relearn 1002026.01 | — | 44.9 | 23.8 | 40 | 16.8 | 84.3 | — | |
| SAREBase Model=mPLUG-Owl-7B, Setting=w/ Relearn 1402026.01 | — | 45.8 | 23.9 | 42 | 17.7 | 84 | — | |
| SAREBase Model=mPLUG-Owl-7B, Setting=w/ LoRA FT2026.01 | — | 47.2 | 25.6 | 46 | 19.3 | 83.8 | — | |
| SAREBase Model=mPLUG-Owl-7B, Setting=w/ Adversarial2026.01 | — | 37.5 | 21.2 | 39 | 14.7 | 71.6 | — | |
| VanillaBase Model=LLaVA-v1.5-7B, Setting=Baseline2026.01 | — | 49.4 | 22.1 | 42 | 14.7 | 85.3 | — | |
| VanillaBase Model=mPLUG-Owl-7B, Setting=Baseline2026.01 | — | 69.6 | 33.5 | 60 | 24.1 | 85.6 | — |