# TTVS: Boosting Self-Exploring Reinforcement Learning via Test-time Variational Synthesis

arXiv: https://arxiv.org/abs/2604.08468 · HTML: https://mlsanity.com/paper/2604.08468/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Placed in Reinforcement learning for reasoning, one of 17 directions in LLM reasoning.
- Builds on 30 of its 48 references that are in the map.
- First posted on arXiv 9 Apr 2026.

## Placed in

- LLM reasoning / [Reinforcement learning for reasoning](https://mlsanity.com/lr/reinforcement-learning-for-reasoning/)
- Deep reinforcement learning methods / [RL training of language and agent models](https://mlsanity.com/deeprl/rl-training-of-language-and-agent-models/)

## Builds on (most cited first, 30 of 30)

- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [Measuring Mathematical Problem Solving With the MATH Dataset](https://mlsanity.com/paper/2103.03874/) (2021)
- [Deep reinforcement learning from human preferences](https://mlsanity.com/paper/1706.03741/) (2017)
- [Constitutional AI: Harmlessness from AI Feedback](https://mlsanity.com/paper/2212.08073/) (2022)
- [GPQA: A Graduate-Level Google-Proof Q&A Benchmark](https://mlsanity.com/paper/2311.12022/) (2023)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [Scaling Laws for Reward Model Overoptimization](https://mlsanity.com/paper/2210.10760/) (2022)
- [Qwen2.5-Math Technical Report: Toward Mathematical Expert Model via Self-Improvement](https://mlsanity.com/paper/2409.12122/) (2024)
- [SimpleRL-Zoo: Investigating and Taming Zero Reinforcement Learning for Open Base Models in the Wild](https://mlsanity.com/paper/2503.18892/) (2025)
- [Open-Reasoner-Zero: An Open Source Approach to Scaling Up Reinforcement Learning on the Base Model](https://mlsanity.com/paper/2503.24290/) (2025)
- [DeepMath-103K: A Large-Scale, Challenging, Decontaminated, and Verifiable Mathematical Dataset for Advancing Reasoning](https://mlsanity.com/paper/2504.11456/) (2025)
- [GPT4Tools: Teaching Large Language Model to Use Tools via Self-instruction](https://mlsanity.com/paper/2305.18752/) (2023)
- [ReFT: Reasoning with Reinforced Fine-Tuning](https://mlsanity.com/paper/2401.08967/) (2024)
- [A Survey on Test-Time Scaling in Large Language Models: What, How, Where, and How Well?](https://mlsanity.com/paper/2503.24235/) (2025)
- [Free Process Rewards without Process Labels](https://mlsanity.com/paper/2412.01981/) (2024)
- [Right Question is Already Half the Answer: Fully Unsupervised LLM Reasoning Incentivization](https://mlsanity.com/paper/2504.05812/) (2025)
- [Toward Self-Improvement of LLMs via Imagination, Searching, and Criticizing](https://mlsanity.com/paper/2404.12253/) (2024)
- [One-Minute Video Generation with Test-Time Training](https://mlsanity.com/paper/2504.05298/) (2025)
- [SeRL: Self-Play Reinforcement Learning for Large Language Models with Limited Data](https://mlsanity.com/paper/2505.20347/) (2025)
- [Synthetic Data Generation & Multi-Step RL for Reasoning & Tool Use](https://mlsanity.com/paper/2504.04736/) (2025)
- [Table-R1: Region-based Reinforcement Learning for Table Understanding](https://mlsanity.com/paper/2505.12415/) (2025)
- [On-Policy RL with Optimal Reward Baseline](https://mlsanity.com/paper/2505.23585/) (2025)
- [DiPrompT: Disentangled Prompt Tuning for Multiple Latent Domain Generalization in Federated Learning](https://mlsanity.com/paper/2403.08506/) (2024)
- [Curriculum-RLAIF: Curriculum Alignment with Reinforcement Learning from AI Feedback](https://mlsanity.com/paper/2505.20075/) (2025)
- [DiEP: Adaptive Mixture-of-Experts Compression through Differentiable Expert Pruning](https://mlsanity.com/paper/2509.16105/) (2025)
- [ReCreate: Reasoning and Creating Domain Agents Driven by Experience](https://mlsanity.com/paper/2601.11100/) (2026)
- [Combating Data Imbalances in Federated Semi-supervised Learning with Dual Regulators](https://mlsanity.com/paper/2307.05358/) (2023)
- [LLM Evaluation Based on Aerospace Manufacturing Expertise: Automated Generation and Multi-Model Question Answering](https://mlsanity.com/paper/2501.17183/) (2025)
- [SeDi-Instruct: Enhancing Alignment of Language Models through Self-Directed Instruction Generation](https://mlsanity.com/paper/2502.04774/) (2025)

## Built on by (most cited first, 1 of 1)

- [Entropy Polarity in Reinforcement Fine-Tuning: Direction, Asymmetry, and Control](https://mlsanity.com/paper/2605.11775/) (2026)

## Abstract (arXiv)

Despite significant advances in Large Reasoning Models (LRMs) driven by reinforcement learning with verifiable rewards (RLVR), this paradigm is fundamentally limited in specialized or novel domains where such supervision is prohibitively expensive or unavailable, posing a key challenge for test-time adaptation. While existing test-time methods offer a potential solution, they are constrained by learning from static query sets, risking overfitting to textual patterns. To address this gap, we introduce Test-Time Variational Synthesis (TTVS), a novel framework that enables LRMs to self-evolve by dynamically augmenting the training stream from unlabeled test queries. TTVS comprises two synergistic modules: (1) Online Variational Synthesis, which transforms static test queries into a dynamic stream of diverse, semantically-equivalent variations, enforcing the model to learn underlying problem logic rather than superficial patterns; (2) Test-time Hybrid Exploration, which balances accuracy-driven exploitation with consistency-driven exploration across synthetic variants. Extensive experiments show TTVS yields superior performance across eight model architectures. Notably, using only unlabeled test-time data, TTVS not only surpasses other test-time adaptation methods but also outperforms state-of-the-art supervised RL-based techniques trained on vast, high-quality labeled data.