AI News

Hao AI Lab launches vstack and veriserve, providing formal verification for coding agents and inference services

UCSD Hao AI LabUpdated Nov 12 03:45

AI summary

Hao AI Lab releases vstack and veriserve: the former is a 5-stage formal verification workflow for coding agents, and the latter is a verifiable PD-disaggregated inference serving engine. The team used vstack to discover 3 classes of RDMA and cache lifecycle defects in a production inference engine and proposed 6 upstream fixes; veriserve's KV page lifecycle design model comes with a TLAPS safety proof comprising 226 modules and 516,973 obligations. When running Qwen3-32B on a 4-GPU NVIDIA GB200 node, the average p50 TPOT overhead introduced by PdLedger is +0.18%, with a 95% CI upper bound of +0.68%.

Why read this

AI score 54

This research proposes a formal verification workflow for coding agents and discovers three classes of RDMA and cache race issues in a production-grade inference engine.

Topics

Tags

Body

This project is done by our amazing @Yuxuan_Zhang13 (lead), @AlexZhang152005 @yisun0618 @Junda_Su08 @Junda_Chen_, and our advisor @haozhangml

Blog: https://t.co/aQiIwcQHj4 vstack: https://t.co/dJ1aFVVCjQ veriserve: https://t.co/tG021LctmW

Source

UCSD Hao AI Lab