Paper 2603.04304
V1: Unifying Generation and Self-Verification for Parallel Reasoners
- Published
- Mar 2026
- Research lab
- Together AI
- Citations
- 7
- GitHub
- 39 stars
01 In brief
Summary
The paper introduces V1, a framework that unifies generation and self-verification for parallel reasoning in LLMs.
It identifies that pointwise self-verification suffers from calibration collapse, while self-aggregation methods like RSA lead to diversity collapse.
V1 comprises two components: V1-Infer, an uncertainty-guided pairwise ranking algorithm using Swiss-system tournament refinement to allocate verification compute to the most uncertain pairs, and V1-PairRL, an RL framework that co-trains a single model as both generator and pairwise self-verifier.
Experiments on code generation (LiveCodeBench, CodeContests, SWE-Bench) and math (AIME, HMMT) show V1-Infer improves Pass@1 by up to 10% over pointwise verification and outperforms RSA with fewer calls.
V1-PairRL achieves 7-9% test-time scaling gains over standard RL and pointwise co-training, and improves base Pass@1 by up to 8.7% in code generation.
The framework demonstrates that pairwise self-verification is more calibrated and diversity-preserving, and that co-training generation and verification yields stronger reasoning models.
02 From the paper
Abstract
Test-time scaling for complex reasoning tasks shows that leveraging inference-time compute, by methods such as independently sampling and aggregating multiple solutions, results in significantly better task outcomes. However, a critical bottleneck is verification: sampling is only effective if correct solutions can be reliably identified among candidates. While existing approaches typically evaluate candidates independently via scalar scoring, we demonstrate that models are substantially stronger at pairwise self-verification. Leveraging this insight, we introduce $V_1$, a framework that unifies generation and verification through efficient pairwise ranking. $V_1$ comprises two components: $V_1$-Infer, an uncertainty-guided algorithm using a tournament-based ranking that dynamically allocates self-verification compute to candidate pairs whose relative correctness is most uncertain; and $V_1$-PairRL, an RL framework that jointly trains a single model as both generator and pairwise self-verifier, ensuring the verifier adapts to the generator's evolving distribution. On code generation (LiveCodeBench, CodeContests, SWE-Bench) and math reasoning (AIME, HMMT) benchmarks, $V_1$-Infer improves Pass@1 by up to $10%$ over pointwise verification and outperforms recent test-time scaling methods while being significantly more efficient. Furthermore, $V_1$-PairRL achieves $7$--$9%$ test-time scaling gains over standard RL and pointwise joint training, and improves base Pass@1 by up to 8.7% over standard RL in a code-generation setting.