Dawn Song's Team Releases Vero Benchmark for AI-Generated Formally Verified Software
dawnsongtweets · x · 2026-08-23
Researchers from UC Berkeley and collaborators released Vero, the first benchmark to evaluate AI agents on joint implementation and proof synthesis at the repository level. Vero includes 43 real-world multi-module instances across domains like cryptography and distributed systems. It features a formal audit mechanism where agents can submit machine-checked proofs to verify specifications or identify errors in reference implementations.
More from coding & agent
- Decision models match LLMs on routing and safety — and are 7-8x faster — randal_olson · 2026-10-16
- Temporal Serverless Workers on Cloud Run: queue-driven autoscaling, scale to zero — rseroter · 2026-10-16
- Meta's Superintelligence Lab Runs 200+ Direct Reports—A Lesson for Flat Multi-Agent Systems — peterjliu · 2026-10-16
- LangChain CEO: the most interesting agent harness work lives at the harness-infra boundary — hwchase17 · 2026-10-16
- GitSwarm paper: agents share branchable Git memory, solving all 30 IMOProofBench problems in one run — anirudhg9119 · 2026-10-16
- Hands-on with Gemini 4 Argon: faster than Opus 5.5 but hallucinates more and costs more — prasenx · 2026-10-16