[2603.19880] What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time

Computer Science > Machine Learning arXiv:2603.19880 (cs) [Submitted on 20 Mar 2026] Title:What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time Authors:Dong Yan, Jian Liang, Yanbo Wang, Shuo Lu, Ran He, Tieniu Tan View a PDF of the paper titled What If Consensus Lies? Selective-Complementary Reinforcement Learning at Test Time, by Dong Yan and 5 other authors View PDF HTML (experimental) Abstract:Test-Time Reinforcement Learning (TTRL) enables Large Language Models (LLMs) to enhance reasoning capabilities on unlabeled test streams by deriving pseudo-rewards from majority voting consensus. However, existing TTRL methods rely exclusively on positive pseudo-labeling strategies. Such reliance becomes vulnerable under challenging scenarios where answer distributions are highly dispersed, resulting in weak consensus that inadvertently reinforces incorrect trajectories as supervision signals. In this paper, we propose SCRL (Selective-Complementary Reinforcement Learning), a robust test-time reinforcement learning framework that effectively mitigates label noise amplification. SCRL develops Selective Positive Pseudo-Labeling, which enforces strict consensus criteria to filter unreliable majorities. Complementarily, SCRL introduces Entropy-Gated Negative Pseudo-Labeling, the first negative supervision mechanism in TTRL, to reliably prune incorrect trajectories based on generation uncertainty. Extensive experiments on multiple reasoning benchmarks demo...