The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning

• URL: http://arxiv.org/abs/2606.29526v1
• Date: Sun, 28 Jun 2026 17:40:02 GMT
• Title: The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning
• Abstract summary: Reinforcement learning (RL) has gained growing attention in large language model (LLM) post-training, yet RL training remains fragile and can suffer from instability or collapse.<n>One vital cause is training-inference mismatch: LLM adopts separate inference and training engines for generation efficiency and training precision.<n>We introduce Monotonic Inference Policy Update (MIPU), a two-step LLM RL framework that constructs sampler-referenced candidate updates and selectively accepts synchronized candidates.
• Score: 65.72688364382331
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reinforcement learning (RL) has gained growing attention in large language model (LLM) post-training, yet RL training remains fragile and can suffer from instability or collapse. One vital cause is training-inference mismatch: LLM adopts separate inference and training engines for generation efficiency and training precision, which in practice exhibits inconsistent probabilities for the same trajectories on training and inference sides, even with synchronized model parameters. This naturally induces a special type of off-policyness ever existing and poisoning the training. Prior works have made various efforts in addressing the off-policyness to stabilize the training policies under the mismatch. In this paper, we point out the objective misalignment neglected by existing works that an effective update to the policy in the training engine not necessarily ensures the improvement of the inference policy, i.e., the one used in deployment. To this end, we propose a new policy optimization objective for LLM RL, named Monotonic Inference Policy Improvement (MIPI). Following this principle, we introduce Monotonic Inference Policy Update (MIPU), a two-step LLM RL framework that constructs sampler-referenced candidate updates and selectively accepts synchronized candidates using an inference-side gap proxy. Experiments conducted on two model scales under high mismatch show that MIPU improves average reasoning performance and training stability.

Related papers

• Uncertainty-Aware LLM-Guided Policy Shaping for Sparse-Reward Reinforcement Learning [2.384534878752428]
  We propose Uncertainty-Aware LLM-Guided Policy Shaping (ULPS), a framework that integrates a calibrated Large Language Model into the Reinforcement Learning training loop.<n>ULPS employs an A*-based oracle to synthesize optimal symbolic trajectories, which are used to fine-tune a BERT-based language model.<n>We observe consistent improvements in success rate, reward efficiency, and sample complexity over unguided, uncalibrated, and standard RL baselines.
  arXiv  Detail & Related papers  (2026-06-04T19:46:45Z)
• Post-Hoc Robustness for Model-Based Reinforcement Learning [1.4072254177584387]
  This work introduces post-hoc robustification of deep RL agents at inference time.<n>The goal is to improve robustness without any additional training of neural networks.
  arXiv  Detail & Related papers  (2026-06-02T11:43:13Z)
• 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)
• A Step Back: Prefix Importance Ratio Stabilizes Policy Optimization [58.116300485427764]
  Reinforcement learning post-training can elicit reasoning behaviors in large language models.<n> token-level correction often leads to unstable training dynamics when the degree of off-policyness is large.<n>We propose a simple yet effective objective, Minimum Prefix Ratio (MinPRO)
  arXiv  Detail & Related papers  (2026-01-30T08:47:19Z)
• Stabilizing Reinforcement Learning with LLMs: Formulation and Practices [61.361819972410046]
  We show why and under what conditions the true sequence-level reward can be optimized via a surrogate token-level objective in policy gradient methods such as REINFORCE.<n>This insight provides a principled explanation for the crucial role of several widely adopted techniques in stabilizing RL training.
  arXiv  Detail & Related papers  (2025-12-01T07:45:39Z)
• Reinforcement Learning with Discrete Diffusion Policies for Combinatorial Action Spaces [57.466101098183884]
  Reinforcement learning (RL) struggles to scale to large, action spaces common in many real-world problems.<n>This paper introduces a novel framework for training discrete diffusion models as highly effective policies in complex settings.
  arXiv  Detail & Related papers  (2025-09-26T21:53:36Z)
• LLM-Guided Reinforcement Learning: Addressing Training Bottlenecks through Policy Modulation [7.054214377609925]
  Reinforcement learning (RL) has achieved notable success in various domains.<n>Training effective policies for complex tasks remains challenging.<n>Existing approaches to mitigate training bottlenecks fall into two categories.
  arXiv  Detail & Related papers  (2025-05-27T03:40:02Z)
• Statistically Efficient Variance Reduction with Double Policy Estimation for Off-Policy Evaluation in Sequence-Modeled Reinforcement Learning [53.97273491846883]
  We propose DPE: an RL algorithm that blends offline sequence modeling and offline reinforcement learning with Double Policy Estimation. We validate our method in multiple tasks of OpenAI Gym with D4RL benchmarks.
  arXiv  Detail & Related papers  (2023-08-28T20:46:07Z)

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.