List your company

Bug injection on MBPP+

99.89Attack Success Rate (ASR)XOXO + GCGS
38.32254.30670.2986.274Mar 18, 2025

Evaluation results

MethodLinks
XOXO + GCGSModel=DeepSeek Coder 6.7B2025.0399.8920
XOXOModel=Llama 3.1 8B2025.0399.8822
XOXO + GCGSModel=Llama 3.1 8B2025.0399.8822
XOXOModel=DeepSeek Coder 6.7B2025.0399.5525
XOXO + GCGSModel=Codestral 22B2025.0399.3637
XOXOModel=Codestral 22B2025.0398.9943
XOXOModel=Claude 3.5 Sonnet v22025.0398.4275
XOXO + GCGSModel=DeepSeek Coder 33B2025.0396.4180
XOXOModel=DeepSeek Coder 33B2025.0395.2784
XOXO + GCGSModel=Qwen 2.5 Coder 7B2025.0384.53169
XOXOModel=Qwen 2.5 Coder 7B2025.0381.29180
XOXO + GCGSModel=Qwen 2.5 Coder 32B2025.0375.37235
XOXOModel=Qwen 2.5 Coder 32B2025.0373.17228
XOXO + GCGSModel=GPT 4o2025.0340.69233