GenPO++: Generative Policy Optimization with Jacobian-free Likelihood Ratios

• URL: http://arxiv.org/abs/2606.06967v1
• Date: Fri, 05 Jun 2026 06:54:09 GMT
• Title: GenPO++: Generative Policy Optimization with Jacobian-free Likelihood Ratios
• Abstract summary: GenPO++ is a reversible generative policy optimization framework.<n>It uses history states as auxiliary memory in a high-order reversible ODE solver.<n>It achieves competitive or superior performance over state-of-the-art on-policy RL methods.
• Score: 39.578640775213906
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Generative policies provide expressive and multimodal action distributions, making them attractive for reinforcement learning (RL) in complex continuous-control tasks. Among them, flow-based policies are especially appealing because they generate actions through deterministic transport maps. However, applying such generative policies to likelihood-based on-policy learning remains limited by the difficulty of evaluating the probability of executed actions. Existing flow RL methods either replace the true action-density ratio with approximate surrogates, which can introduce biased updates, or recover exact likelihoods through dummy-action augmentation, which enlarges the policy space and increases computation. In this work, we propose GenPO++, a reversible generative policy optimization framework that uses history states as auxiliary memory in a high-order reversible ODE solver, yielding exact inversion without changing the original action dimension. The resulting generative policy map has a log-determinant determined only by fixed solver coefficients, enabling exact and Jacobian-free likelihood-ratio computation. This design preserves the expressiveness of generative flow policies while avoiding both action ratio bias and dummy-action overhead. We evaluate GenPO++ on large-scale simulated control, fine-tuning, and real-world robotic manipulation tasks, where it achieves competitive or superior performance over state-of-the-art on-policy RL methods, while improving training stability and computational efficiency.

Related papers

• One-Way Policy Optimization for Self-Evolving LLMs [63.8638342097375]
  Reinforcement Learning with Verifiable Rewards (RLVR) has become a promising paradigm for scaling reasoning capabilities of Large Language Models (LLMs)<n>We propose One-Way Policy Optimization (OWPO), a method based on the principle of decoupling optimization direction from update magnitude.<n> Experimental results demonstrate that OWPO outperforms strong baselines, including DAPO, OPD, and MOPD.
  arXiv  Detail & Related papers  (2026-05-21T08:25:27Z)
• Stochastic MeanFlow Policies: One-Step Generative Control with Entropic Mirror Descent [53.828537014796574]
  Online off-policy reinforcement learning (RL) is shaped by two choices: the policy class and the update rule.<n>We propose MeanFlow Policies, a one-step generative policy class that maps noise to actions through a MeanFlow transformation.<n>Across seven MuJoCo benchmarks, S improves over Gaussian and generative baselines while retaining single-step inference efficiency.
  arXiv  Detail & Related papers  (2026-05-20T15:14:14Z)
• Dichotomous Diffusion Policy Optimization [46.51375996317989]
  DIPOLE is a novel RL algorithm designed for stable and controllable diffusion policy optimization.<n>We also use DIPOLE to train a large vision-language-action model for end-to-end autonomous driving.
  arXiv  Detail & Related papers  (2025-12-31T16:56:56Z)
• GoRL: An Algorithm-Agnostic Framework for Online Reinforcement Learning with Generative Policies [16.859964356466676]
  GoRL is a framework that optimize a tractable latent policy while utilizing a conditional generative decoder to synthesize actions.<n>GoRL consistently outperforms both Gaussian policies and recent generative-policy baselines.
  arXiv  Detail & Related papers  (2025-12-02T09:49:26Z)
• Random Policy Valuation is Enough for LLM Reasoning with Verifiable Rewards [47.557539197058496]
  We introduce Random Policy Valuation for Diverse Reasoning (ROVER)<n>ROVER is a minimalist yet highly effective RL method that samples actions from a softmax over uniform-policy Q-values.<n>It demonstrates superior performance in both textbfquality (textbf+8.2 on pass@1, textbf+16.8 on pass@256) and textbfdiversity (textbf+17.6%)
  arXiv  Detail & Related papers  (2025-09-29T16:09:07Z)
• GenPO: Generative Diffusion Models Meet On-Policy Reinforcement Learning [34.25769740497309]
  GenPO is a generative policy optimization framework that leverages exact diffusion inversion to construct invertible action mappings.<n>GenPO is the first method to successfully integrate diffusion policies into on-policy RL, unlocking their potential for large-scale parallelized training and real-world robotic deployment.
  arXiv  Detail & Related papers  (2025-05-24T15:57:07Z)
• Dense Policy: Bidirectional Autoregressive Learning of Actions [51.60428100831717]
  This paper introduces a bidirectionally expanded learning approach, termed Dense Policy, to establish a new paradigm for autoregressive policies in action prediction.<n>It employs a lightweight encoder-only architecture to iteratively unfold the action sequence from an initial single frame into the target sequence in a coarse-to-fine manner.<n>Experiments validate that our dense policy has superior autoregressive learning capabilities and can surpass existing holistic generative policies.
  arXiv  Detail & Related papers  (2025-03-17T14:28:08Z)
• REBEL: Reinforcement Learning via Regressing Relative Rewards [59.68420022466047]
  We propose REBEL, a minimalist RL algorithm for the era of generative models.<n>In theory, we prove that fundamental RL algorithms like Natural Policy Gradient can be seen as variants of REBEL.<n>We find that REBEL provides a unified approach to language modeling and image generation with stronger or similar performance as PPO and DPO.
  arXiv  Detail & Related papers  (2024-04-25T17:20:45Z)
• Offline Reinforcement Learning with Closed-Form Policy Improvement Operators [88.54210578912554]
  Behavior constrained policy optimization has been demonstrated to be a successful paradigm for tackling Offline Reinforcement Learning. In this paper, we propose our closed-form policy improvement operators. We empirically demonstrate their effectiveness over state-of-the-art algorithms on the standard D4RL benchmark.
  arXiv  Detail & Related papers  (2022-11-29T06:29:26Z)
• PFPN: Continuous Control of Physically Simulated Characters using Particle Filtering Policy Network [0.9137554315375919]
  We propose a framework that considers a particle-based action policy as a substitute for Gaussian policies. We demonstrate the applicability of our approach on various motion capture imitation tasks.
  arXiv  Detail & Related papers  (2020-03-16T00:35:36Z)

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.