Create

Sign in to ReadmeX

or

How we handle your data: Privacy Policy

ReadmeX

Your AI reading companion

What would you like to catch up on?

Pick a briefing, or ask your own question.

Your community and people briefings stay personal to you.

Story

AuXSpec trains Lean spec generators with execution-based rewards

AI summary

Haocheng Wang and six coauthors submitted the AuXSpec paper to arXiv on November 16, 2026. AuXSpec is a Lean framework that pairs declarative postconditions with executable counterparts, allowing test cases and execution agreement to provide reinforcement-learning feedback for specification generation; the authors also synthesized 3,037 LeetCode problems and expanded training to 19.5K samples through expert iteration on Codeforces. The paper reports that training Qwen3.5-9B and Qwen3.8-27B with this reward raised AuXSpec-Test pass@4 to 82.7% and 84.3%, respectively, and says the authors released the trained weights, synthesized problems, and the expert-reviewed AuXSpec-Test benchmark.

In-depth insight

AuXSpec offers an execution-based way to train Lean specification generators without requiring a separate explicit proof for every training signal, and the authors report releasing both model weights and evaluation data.

Computer Science > Software Engineering

arXiv:2611.21421 (cs)

[Submitted on 16 Nov 2026]

Title:AuXSpec: Learning to Autoformalize Program Specifications via Executable Postconditions

Authors:Haocheng Wang, Weixuan Yuan, Florian Hübler, Yinya Huang, Zhijiang Guo, Sorrachai Yingchareonthawornchai, Rasmus Kyng

View a PDF of the paper titled AuXSpec: Learning to Autoformalize Program Specifications via Executable Postconditions, by Haocheng Wang and 6 other authors

#AuXSpec#Lean#Qwen3.5-9B#Qwen3.8-27B#AuXSpec-Test#Verina#VeriSpecGen-30B-A3B#LeetCode#Codeforces

Original · 3 min readRead the original
Group chat

No comments yet. Start the conversation.