Skip to content
Preprint

Le Critique: Privileged Value Functions for LLM Reinforcement Learning

Aug 2026 · 0 citations · 42 references
Computer Science

TL;DR

This work proposes two complementary strategies to improve the performance of value function RL: Privileged Value Functions (PVF) which provide an elegant mechanism to inject additional task-relevant token-level signal without biasing the policy objective; and TETHER, a baseline that adaptively interpolates between group-relative and value baselines depending on the value function accuracy.

Abstract

Reinforcement learning algorithms for Large Language Models (LLMs) are largely distinguished by their variance reduction strategy. Group-relative methods like GRPO reduce gradient variance by sampling multiple rollouts per prompt, but provide only sequence-level credit. Training is also blocked by straggler rollouts, reducing throughput and increasing off-policyness. Learned value functions theoretically address both problems, providing token-level advantages without requiring large groups. However, additional infrastructure engineering challenges combined with the practical success of critic-free methods have made it difficult to justify their inclusion in RL pipelines. We propose two complementary strategies to improve the performance of value function RL: 1) Privileged Value Functions (PVF) which provide an elegant mechanism to inject additional task-relevant token-level signal without biasing the policy objective; 2) TETHER, a baseline that adaptively interpolates between group-relative and value baselines depending on the value function accuracy. Across several reasoning tasks, both strategies consistently improve over the standard value function baseline, and are competitive with or outperform mean-baseline GRPO.

View source

Similar papers

Preprint Aug 2026

SAPO: Single-Rollout Autoregressive Policy Optimization for Agentic Reinforcement Learning

This work proposes Single-rollout Autoregressive Policy Optimization (SAPO), a low-memory and compute-efficient framework in which the policy and value functions share a single autoregressive backbone, and introduces a trajectory-level generalized advantage estimator that combines lambda-returns with batch normalization.

D. Liang, Lang Feng, Bo An et al. · 1 citation
Preprint Aug 2026

CVPO: Enhancing LLM Reinforcement Learning Reasoning via Value-Variance Adaptation and Dynamic Curriculum Learning

Reinforcement learning (RL) has emerged as an effective method for enhancing the reasoning capabilities of large language models (LLMs). However, existing methods suffer from insufficient precision in feedback on generated answer trajectories and exhibit the phenomenon of problem difficulty drift. To address these challenges, we propose CVPO - Curriculum-guided Value-Variance Policy Optimization. At the response trajectory level, we find that token-level value-variance correlates with exploration intensity. Our theoretical analysis shows this variance bounds policy update magnitude. We then use the estimated trajectory value-variance to quantify the intrinsic randomness in generation. Based on this, we design a variance-aware advantage adjustment mechanism for different reward types. At the question level, we introduce a dynamic curriculum weighting method that adapts to question difficulty. This helps the model focus on tasks matched to its current ability during each training stage. Experimental results show our method outperforms strong value-based baselines like VAPO. It achieves better performance and stronger exploration, enabling more accurate and robust reasoning in language models across various math tasks.

Ziqi Jia, Yalu Ouyang, Bo Pang et al. · 0 citations
Preprint Jul 2026

ARMOR: Stabilizing On-Policy LLM RL with Off-Policy Anchor Samples

This work proposes ARMOR (Anchor Rollout and Mixed Optimization for RL), a framework that shifts the paradigm from passive penalty to active sample stabilization, enabling sustained performance improvements over extended training horizons.

Kexin Huang, Junkang Wu, Jinda Lu et al. · 0 citations
Preprint Aug 2026

Best Practice Critic Optimization

Best Practice Critic Optimization (BPCO) is developed, a recipe that combines DPPO, value predictions bounded to the reward range, Monte Carlo value targets, unnormalized policy advantages, and length-adaptive generalized advantage estimation and shows that a carefully designed critic provides a reliable alternative to group-relative advantage estimation.

Penghui Qi, Xiangxin Zhou, W. Lee · 0 citations
Preprint Jul 2026

Learning as Reasoning Unfolds: Progressive Rollout Allocation for Efficient Reinforcement Learning

VarIance Guided Online Rollout allocation (VIGOR) is proposed which instead of allocating a fixed rollout budget per example, begins with a small number of rollouts for all examples in a batch and iteratively allocates additional rollouts to those with the highest group reward variance until a fixed total rollout budget is reached.

Heyang Jiang, Henry Liu, Baharan Mirzasoleiman · 1 citation