EG-CFG (DeepSeek-V3-0324)

Reported on 5 benchmarks across 1 task · 1 paper · 4 SOTA

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Natural Language Processing5 results

Code GenerationonHumanEval-ET
Pass@1· 2025-06-12
87.19
SOTA
Execution Guided Line-by-Line Code Generation arXiv:2506.10948
Code GenerationonCodeContests
Test Set pass@1· 2025-06-12
58.18
SOTA
Execution Guided Line-by-Line Code Generation arXiv:2506.10948
Code GenerationonMBPP
Accuracy· 2025-06-12
96.6
SOTA
Execution Guided Line-by-Line Code Generation arXiv:2506.10948
Code GenerationonMBPP-ET
Pass@1· 2025-06-12
73
SOTA
Execution Guided Line-by-Line Code Generation arXiv:2506.10948
Code GenerationonHumanEval
Pass@1· 2025-06-12
96.95
best: 100 (DeepSeek-R1 (MGDebugger))
Execution Guided Line-by-Line Code Generation arXiv:2506.10948