Reproduce “Veri-Sure: Multi-Agent RTL Code Generation with Temporal Tracing, Slicing and Formal Verification” on OpenResearch
Turn your favorite coding agents into research agents that review literature, develop hypotheses, and run experiments.
Universal app for Apple silicon and Intel. Requires macOS 11 or later.
About this paper
Jiale Liu, Taiyu Zhou, Tianqi Jiang
Veri-Sure generates RTL from a hardware specification, then uses a shared contract, simulation traces, dependency slicing, and formal-checking hints to localize repairs to suspect logic. On 209 VerilogEval-v2-EXT tasks, GPT-5.2 with Veri-Sure achieves 93.30% functional Pass@1 overall and 85.07% on Hard tasks. These results measure golden-testbench passes, not exhaustive hardware correctness.
Summary by alphaXiv.
Tip: change arxiv.org to autoarxiv.org in any arXiv abstract link to open the paper here.