Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Question Answering on PIQA (Attack Robustness Evaluation)

77.31Accuracy (Baseline)

Hymba-1.5b

66.618869.394472.1774.9456Dec 14, 2025
Updated 4d ago

Evaluation Results

MethodLinks
2025.12
77.312.67
2025.12
74.8617.33
2025.12
74.4852
2025.12
73.990
2025.12
73.3418
2025.12
73.2925.33
2025.12
72.476
2025.12
72.250
2025.12
71.7649.33
2025.12
68.7724
2025.12
67.0344.66