DyCodeEval (ICML 2025) enables dynamic benchmarking for code LLMs. This collection features dynamic HumanEval and MBPP sets generated with Claude 3.5.
Simin Chen
CM
AI & ML interests
None yet
Organizations
datasets 12
CM/humaneval_trans_java_python
Viewer
• Updated
• 164 • 48
CM/Dynamic_LeetCode
Viewer
• Updated
• 2.87k • 105
CM/Dynamic_MBPP_sanitized
Viewer
• Updated
• 15.8k • 10
CM/Dynamic_HumanEvalZero
Viewer
• Updated
• 15.7k • 15
CM/codexglue_codetrans
Viewer
• Updated
• 11.8k • 20 • 2
CM/codexglue_code2text_ruby
Viewer
• Updated
• 27.6k • 1.96k • 1
CM/codexglue_code2text_python
Viewer
• Updated
• 281k • 2.84k • 8
CM/codexglue_code2text_php
Viewer
• Updated
• 268k • 1.93k • 2
CM/codexglue_code2text_javascript
Viewer
• Updated
• 65.2k • 2.24k • 12
CM/codexglue_code2text_java
Viewer
• Updated
• 181k • 2.46k • 4