| multiple-choice-question-answering-mcqa-on-arc-c | iLLaDA 8B Base | #3 | Accuracy: 60.8 |
| multi-task-language-understanding-on-mmlu | iLLaDA 8B Base | #19 | Average (%): 74.8 |
| multi-task-language-understanding-on-mmlu | iLLaDA 8B Instruct | #25 | Average (%): 71.6 |
| multi-task-language-understanding-on-mmlu-redux | iLLaDA 8B Instruct | #31 | Accuracy: 76.4 |
| math-word-problem-solving-on-math | iLLaDA 8B Instruct | #64 | Accuracy: 56.7 |
| math-word-problem-solving-on-math | iLLaDA 8B Base | #122 | Accuracy: 38.4 |
| question-answering-on-piqa | Confidence-based scoring | #67 | Accuracy: 78.5 |
| question-answering-on-piqa | iLLaDA 8B Base | #68 | Accuracy: 78.5 |
| arithmetic-reasoning-on-gsm8k | iLLaDA 8B Instruct | #82 | Accuracy: 89.0 |
| arithmetic-reasoning-on-gsm8k | iLLaDA 8B Base | #142 | Accuracy: 81.9 |
| sentence-completion-on-hellaswag | iLLaDA 8B Base | #85 | Accuracy: 76.6 |
| code-generation-on-humaneval | iLLaDA 8B Instruct | #91 | Pass@1: 65.9 |
| code-generation-on-humaneval | iLLaDA 8B Base | #101 | Pass@1: 50 |
| code-generation-on-mbpp | iLLaDA 8B Instruct | #91 | Accuracy: 58 |
| code-generation-on-mbpp | iLLaDA 8B Base | #92 | Accuracy: 57.8 |
| multi-task-language-understanding-on-mmlu-pro | iLLaDA 8B Instruct | #111 | Accuracy: 52.3 |