$V_{0.5}$: Generalist Value Model as a Prior for Sparse RL Rollouts

• URL: http://arxiv.org/abs/2603.10848v1
• Date: Wed, 11 Mar 2026 14:57:41 GMT
• Title: $V_{0.5}$: Generalist Value Model as a Prior for Sparse RL Rollouts
• Abstract summary: Generalist Value Models (such as $V_0.5$) achieve pre-trained value estimation by explicitly encoding model capabilities in-context.<n>We propose $V_0.5$, which adaptively fuses the baseline predicted by such value model with the empirical mean derived from sparse rollouts.<n>$V_0.5$ significantly outperforms GRPO and DAPO, achieving faster convergence and over some 10% performance improvement.
• Score: 81.48669089692189
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: In Reinforcement Learning with Verifiable Rewards (RLVR), constructing a robust advantage baseline is critical for policy gradients, effectively guiding the policy model to reinforce desired behaviors. Recent research has introduced Generalist Value Models (such as $V_0$), which achieve pre-trained value estimation by explicitly encoding model capabilities in-context, eliminating the need to synchronously update the value model alongside the policy model. In this paper, we propose $V_{0.5}$, which adaptively fuses the baseline predicted by such value model (acting as a prior) with the empirical mean derived from sparse rollouts. This constructs a robust baseline that balances computational efficiency with extremely low variance. Specifically, we introduce a real-time statistical testing and dynamic budget allocation. This balances the high variance caused by sparse sampling against the systematic bias (or hallucinations) inherent in the value model's prior. By constructing a hypothesis test to evaluate the prior's reliability in real-time, the system dynamically allocates additional rollout budget on demand. This mechanism minimizes the baseline estimator's Mean Squared Error (MSE), guaranteeing stable policy gradients, even under extreme sparsity with a group size of 4. Extensive evaluations across six mathematical reasoning benchmarks demonstrate that $V_{0.5}$ significantly outperforms GRPO and DAPO, achieving faster convergence and over some 10% performance improvement.

Related papers

• CollabEval: Statistically Efficient Collaborative Model Evaluation via Matrix Completion [65.53153291902454]
  Collaborative Evaluation (CollabEval) is a simple, effective, and principled method for exploiting dependencies between historical runs of different models on the same tasks to improve statistical efficiency.<n>We build a low-rank approximation of the score matrix, and use the reconstructed values to guarantee unbiased estimates of the true evaluation metric mean.<n>We find that CollabEval substantially reduces the mean confidence interval size, and the mean squared error of the point estimate, compared to baseline methods at the same annotation budget.
  arXiv  Detail & Related papers  (2026-07-06T13:22:04Z)
• Ratio-Variance Regularized Policy Optimization [64.95520246570446]
  We show that explicitly constraining the policy ratio variance provides a principled local approximation to trust-region constraints.<n>We introduce $bf R2bf VPO$ (Ratio-Variance Regularized Policy Optimization), which implements this constraint via a primal-dual optimization framework.
  arXiv  Detail & Related papers  (2026-05-26T09:53:42Z)
• Your Language Model is Its Own Critic: Reinforcement Learning with Value Estimation from Actor's Internal States [14.658333957936769]
  We introduce Policy Optimization with Internal State Value Estimation.<n>It obtains a baseline at negligible cost by using the policy model's internal signals already computed during the policy forward pass.<n>A lightweight probe predicts the expected verifiable reward from the hidden states of the prompt and generated trajectory.
  arXiv  Detail & Related papers  (2026-05-08T10:49:36Z)
• Unbiased Dynamic Pruning for Efficient Group-Based Policy Optimization [60.87651283510059]
  Group Relative Policy Optimization (GRPO) effectively scales LLM reasoning but incurs prohibitive computational costs.<n>We propose Dynamic Pruning Policy Optimization (DPPO), a framework that enables dynamic pruning while preserving unbiased gradient estimation.<n>To mitigate the data sparsity induced by pruning, we introduce Dense Prompt Packing, a window-based greedy strategy.
  arXiv  Detail & Related papers  (2026-03-04T14:48:53Z)
• Stable Asynchrony: Variance-Controlled Off-Policy RL for LLMs [19.079556051442168]
  Reinforcement learning (RL) is widely used to improve large language models on reasoning tasks.<n>But for widely adopted critic-free policy-gradient methods such as REINFORCE and GRPO, high asynchrony makes the policy-gradient estimator markedly noisy.<n>We propose a stabilization method for REINFORCE/ GRPO-style algorithms that scales learning rate based on effective sample size to dampen unreliable updates.
  arXiv  Detail & Related papers  (2026-02-19T18:40:51Z)
• Fault-Tolerant Evaluation for Sample-Efficient Model Performance Estimators [13.227055178509524]
  We propose a fault-tolerant evaluation framework that integrates bias and variance considerations within an adjustable tolerance level.<n>We show that proper calibration of $varepsilon$ ensures reliable evaluation across different variance regimes.<n> Experiments on real-world datasets demonstrate that our framework provides comprehensive and actionable insights into estimator behavior.
  arXiv  Detail & Related papers  (2026-02-06T22:14:46Z)
• Rethinking the Trust Region in LLM Reinforcement Learning [72.25890308541334]
  Proximal Policy Optimization (PPO) serves as the de facto standard algorithm for Large Language Models (LLMs)<n>We propose Divergence Proximal Policy Optimization (DPPO), which substitutes clipping with a more principled constraint.<n>DPPO achieves superior training and efficiency compared to existing methods, offering a more robust foundation for RL-based fine-tuning.
  arXiv  Detail & Related papers  (2026-02-04T18:59:04Z)
• $V_0$: A Generalist Value Model for Any Policy at State Zero [80.7505802128501]
  Policy methods rely on a baseline to measure the relative advantage of an action.<n>This baseline is typically estimated by a Value Model (Critic) often as large as the policy model itself.<n>We propose a Generalist Value Model capable of estimating the expected performance of any model on unseen prompts.
  arXiv  Detail & Related papers  (2026-02-03T14:35:23Z)
• Ratio-Variance Regularized Policy Optimization for Efficient LLM Fine-tuning [48.34492357368989]
  We propose a primal-dual framework that supports stable on-policy learning and enables principled off-policy data reuse.<n>$R2VPO$ achieves superior performance with average relative gains of up to 17% over strong clipping-based baselines.
  arXiv  Detail & Related papers  (2026-01-06T14:01:42Z)
• Conservative Bayesian Model-Based Value Expansion for Offline Policy Optimization [41.774837419584735]
  offline reinforcement learning (RL) addresses the problem of learning a performant policy from a fixed batch of data collected by following some behavior policy. Model-based approaches are particularly appealing since they can extract more learning signals from the logged dataset by learning a model of the environment.
  arXiv  Detail & Related papers  (2022-10-07T20:13:50Z)
• COMBO: Conservative Offline Model-Based Policy Optimization [120.55713363569845]
  Uncertainty estimation with complex models, such as deep neural networks, can be difficult and unreliable. We develop a new model-based offline RL algorithm, COMBO, that regularizes the value function on out-of-support state-actions. We find that COMBO consistently performs as well or better as compared to prior offline model-free and model-based methods.
  arXiv  Detail & Related papers  (2021-02-16T18:50:32Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.