List your company

Distractor Generation on CLOTH Hard

4.2Invalid RatioQwen 2.5 7B
3.6927.12110.5513.979Nov 3, 2025

Evaluation results

MethodLinks
Qwen 2.5 7Btraining=DCDG-trained2025.114.2
Gemma 2 9Btraining=DCDG-trained2025.115.1
Llama 3.1 8Btraining=DCDG-trained2025.115.2
GPT-4oprompting=5-shot2025.116.8
GPT-4oprompting=0-shot2025.1116.9