AI ResearchJun 30, 2026, 5:39 PM

AxDafny: Agentic Verified Code Generation in Dafny

TickrWire Editorial Desk·Jun 30, 2026, 5:39 PM·1 min read AI-assisted, human-reviewed

Reported by arXiv cs.AI: AxDafny: Agentic Verified Code Generation in Dafny. Analysis and context written by TickrWire.

30-second summary

AxDafny introduces a verifier-guided repair framework for agentic code generation in Dafny, improving verification success over GPT-5.5 on a new benchmark (LCB-Pro-Dafny) of 250 formal programming problems.

TickrWire
Full story

We study agentic code generation in Dafny, where a model must generate both executable code and the proof artifacts for verification. We present AxDafny, a verifier-guided repair framework that iteratively generates implementations, invariants, assertions, and termination arguments. We also introduce LiveCodeBench-Pro-Dafny (LCB-Pro-Dafny), a benchmark of 250 competition-style programming problems translated into Dafny with formal specifications and a verifier-based evaluation harness. On LCB-Pro-Dafny, AxDafny substantially improves verification success over baseline GPT-5.5 performance. On D

Sources · 1
Read next
More stories