> ## Documentation Index
> Fetch the complete documentation index at: https://rllm-org-rllm-19-feat-renderer-parser-backend.mintlify.site/llms.txt
> Use this file to discover all available pages before exploring further.

# V1: Parallel Self-Verification

> Unifying generation and pairwise self-verification for parallel reasoners

<Card title="Paper" icon="file-lines" href="https://arxiv.org/abs/2603.04304">
  arXiv:2603.04304
</Card>

V1 is a framework that improves how models verify multiple solution candidates during inference. Instead of scoring solutions individually, V1 leverages pairwise self-verification — where models compare two candidates head-to-head — combined with a tournament-based ranking algorithm to efficiently allocate verification compute. The training method jointly develops generation and verification capabilities, achieving improvements of up to 10% on code generation and math reasoning benchmarks.