Skip to content

AuXSpec: Learning Autoformalized Program Specifications via Executable Postconditions

AuXSpec: Learning to Autoformalize Program Specifications via Executable Postconditions

Hugging Face Daily Papers· arXiv:2611.21421· Haocheng Wang, Weixuan Yuan, Florian Hübler, Yinya Huang, Zhijiang Guo, Sorrachai Yingchareonthawornchai, Rasmus Kyng·Indexed by this site · 1 days ago· Originally publishedAI updates· Its place in the scan record →
AI summary

AuXSpec pairs declarative postconditions with executable versions, using the consistency of test case execution as the reinforcement learning reward. Using 3,037 LeetCode problems and Codeforces expert iteration to expand to 19.5K samples, Qwen3.5-9B and Qwen3.8-27B achieve SpecGen pass@4 of 82.7% and 84.3% on AuXSpec-Test, with the 27B model surpassing all frontier models.

Read originalhuggingface.co