# Beyond Majority Voting: Towards Fine-grained and More Reliable Reward Signal for Test-Time Reinforcement Learning

arXiv: https://arxiv.org/abs/2512.15146 · HTML: https://mlsanity.com/paper/2512.15146/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 26 of its 45 references that are in the map.
- First posted on arXiv 17 Dec 2025.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)

## Builds on (most cited first, 26 of 26)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models](https://mlsanity.com/paper/2402.03300/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [Evaluating Large Language Models Trained on Code](https://mlsanity.com/paper/2107.03374/) (2021)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Let's Verify Step by Step](https://mlsanity.com/paper/2305.20050/) (2023)
- [HybridFlow: A Flexible and Efficient RLHF Framework](https://mlsanity.com/paper/2409.19256/) (2024)
- [Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement](https://mlsanity.com/paper/2409.12122/) (2024)
- [Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning](https://mlsanity.com/paper/2506.01939/) (2025)
- [SimpleRL-Zoo: Investigating and Taming Zero Reinforcement Learning for Open Base Models in the Wild](https://mlsanity.com/paper/2503.18892/) (2025)
- [Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learning](https://mlsanity.com/paper/1908.02983/) (2019)
- [Reinforcement Learning for Reasoning in Large Language Models with One Training Example](https://mlsanity.com/paper/2504.20571/) (2025)
- [Reinforcement Learning with Verifiable Rewards Implicitly Incentivizes Correct Reasoning in Base LLMs](https://mlsanity.com/paper/2506.14245/) (2025)
- [Learning to Reason without External Rewards](https://mlsanity.com/paper/2505.19590/) (2025)
- [Crossing the Reward Bridge: Expanding RL with Verifiable Rewards Across Diverse Domains](https://mlsanity.com/paper/2503.23829/) (2025)
- [AdaCoT: Pareto-Optimal Adaptive Chain-of-Thought Triggering via Reinforcement Learning](https://mlsanity.com/paper/2505.11896/) (2025)
- [RLPR: Extrapolating RLVR to General Domains without Verifiers](https://mlsanity.com/paper/2506.18254/) (2025)
- [Panacea: Pareto Alignment via Preference Adaptation for LLMs](https://mlsanity.com/paper/2402.02030/) (2024)
- [SENTRY: Selective Entropy Optimization via Committee Consistency for Unsupervised Domain Adaptation](https://mlsanity.com/paper/2012.11460/) (2020)
- [Evolving Language Models without Labels: Majority Drives Selection, Novelty Promotes Variation](https://mlsanity.com/paper/2509.15194/) (2025)
- [Beyond Correctness: Harmonizing Process and Outcome Rewards through RL Training](https://mlsanity.com/paper/2509.03403/) (2025)
- [Towards High Data Efficiency in Reinforcement Learning with Verifiable Reward](https://mlsanity.com/paper/2509.01321/) (2025)
- [Post-Training Large Language Models via Reinforcement Learning from Self-Feedback](https://mlsanity.com/paper/2507.21931/) (2025)
- [RLAC: Reinforcement Learning with Adversarial Critic for Free-Form Generation Tasks](https://mlsanity.com/paper/2511.01758/) (2025)
- [The Self-Improvement Paradox: Can Language Models Bootstrap Reasoning Capabilities without External Scaffolding?](https://mlsanity.com/paper/2502.13441/) (2025)
- [Improving Rationality in the Reasoning Process of Language Models through Self-playing Game](https://mlsanity.com/paper/2506.22920/) (2025)

## Built on by (most cited first, 13 of 13)

- [Dynamics Within Latent Chain-of-Thought: An Empirical Study of Causal Structure](https://mlsanity.com/paper/2602.08783/) (2026)
- [Detecting and Mitigating the Correct-Answer Extinction Window in Test-Time Reinforcement Learning with Majority Voting](https://mlsanity.com/paper/2605.19444/) (2026)
- [When Self-Belief Misleads: Active Label Acquisition for Reinforcement Learning with Verifiable Rewards](https://mlsanity.com/paper/2605.25864/) (2026)
- [Distribution-Aware Reward Estimation for Test-Time Reinforcement Learning](https://mlsanity.com/paper/2601.21804/) (2026)
- [SAT: Balancing Reasoning Accuracy and Efficiency with Stepwise Adaptive Thinking](https://mlsanity.com/paper/2604.07922/) (2026)
- [ConsistRM: Improving Generative Reward Models via Consistency-Aware Self-Training](https://mlsanity.com/paper/2604.07484/) (2026)
- [Hi-TTRL: Regulating Consensus with Hints for Test-Time Reinforcement Learning](https://mlsanity.com/paper/2608.03545/) (2026)
- [From the Inside Out: Progressive Distribution Refinement for Confidence Calibration](https://mlsanity.com/paper/2603.16500/) (2026)
- [Unsupervised Post-Training of Foundation Models: A Survey](https://mlsanity.com/paper/2608.24982/) (2026)
- [Harnessing Image Question Dependence for Better VLM Test-time Reinforcement Learning](https://mlsanity.com/paper/2609.13296/) (2026)
- [Test-time Reinforcement Learning for Anomalous Video Understanding](https://mlsanity.com/paper/2609.26099/) (2026)
- [Group-Marginalized Self-Rewarding RL Drives Zero-Label Self-Evolving](https://mlsanity.com/paper/2609.36750/) (2026)
- [SelectTTRL: Representation-Guided Data Selection for Efficient Test-Time Reinforcement Learning](https://mlsanity.com/paper/2610.16104/) (2026)

## Abstract (arXiv)

Test-time reinforcement learning mitigates the reliance on annotated data by using majority voting results as pseudo-labels, emerging as a complementary direction to reinforcement learning with verifiable rewards (RLVR) for improving reasoning ability. However, this voting strategy often induces confirmation bias and suffers from sparse rewards, limiting the overall performance. In this work, we propose subgroup-specific step-wise confidence-weighted pseudo-label estimation (SCOPE), a framework integrating model confidence and dynamic subgroup partitioning to address these issues. Specifically, SCOPE integrates the proposed step-wise confidence into pseudo label estimation, prioritizing high-quality reasoning paths over simple frequency count. Furthermore, it dynamically partitions the candidate outputs pool into independent subgroups by balancing reasoning quality against exploration diversity. By deriving local consensus via repeat sampling for each sub group, SCOPE provides diverse supervision targets to encourage broader exploration. We conduct experiments across various models and benchmarks, experimental results show that SCOPE consistently outperforms recent baselines. Notably, SCOPE achieving relative improvements of 13.1% on challenging AIME 2025 and 8.1% on AMC. The code is released at https://github.com/szu-tera/SCOPE.