List your company

Adversarial Robustness Evaluation on ASAG

1.3ASRPAIR
-2.64824.00150.6577.299Jun 2, 2026

Evaluation results

MethodLinks
PAIRGrader=Claude-Sonnet-4.62026.061.3-0.0250.135
DANGrader=Claude-Sonnet-4.62026.068.20.0190.267
PAIRGrader=GPT-5-mini2026.0628.80.030.096
MSJGrader=Gemini-3-flash2026.06500.0580.001
PAIRGrader=Qwen3-4B2026.06500.1710
MSJGrader=Claude-Sonnet-4.62026.0653.10.1580
PAIRGrader=Gemini-3-flash2026.0655.60.1110
MSJGrader=GPT-5-mini2026.0657.10.0640.002
MSJGrader=Qwen3-4B2026.0659.40.1840
PAIRGrader=Llama-3.1-8B2026.0660.80.2380
DANGrader=GPT-5-mini2026.0668.30.1220
DANGrader=Gemini-3-flash2026.0684.10.1960
DANGrader=Qwen3-4B2026.0691.90.3410
DANGrader=Llama-3.1-8B2026.061000.4980
MSJGrader=Llama-3.1-8B2026.061000.4520