List your company

Tool Selection on ScienceQA

99.29P(Use)Vicuna-7B
1.42626.83352.2477.647Apr 7, 2025

Evaluation results

MethodLinks
Vicuna-7BStatus=Attack (SEEM)2025.0499.291,812
Llama-8BStatus=Attack (SEEM)2025.0495.28173
Llama-3.2-3BStatus=Attack (SEEM)2025.0470.01172
GPT-3.5Status=Attack (SEEM)2025.044624
Qwen-2.5-7BStatus=Attack (SEEM)2025.0442.7865
Phi-4-14BStatus=Attack (SEEM)2025.0441.16109
GPT-3.5Status=Origin2025.0437—
Llama-8BStatus=Origin2025.0434.91—
GPT-4oStatus=Attack (SEEM)2025.0429.8834
Qwen-2.5-7BStatus=Origin2025.0425.94—
Llama-3.2-3BStatus=Origin2025.0425.71—
GPT-4oStatus=Origin2025.0422.29—
Phi-4-14BStatus=Origin2025.0419.71—
Vicuna-7BStatus=Origin2025.045.19—