github link : genji970/Scaling-Test-Time-Compute-for-Agentic-Coding-: paper implementation of Meta Ai
paper link : https://arxiv.org/abs/2604.16529v1
As far as I know, there is no public implementation of this paper yet, so I built a minimal research implementation of the core PDR+RTV pipeline.
I made project to run gemini-3.1-pro model and test on SWE benchmark(In paper, there is one more benchmark and used models such as opus and more)
Need gemini-api-key to run.
[link] [comments]