CORTEXA
← Browse
arxivcs.AI2026-07-06

Diffusion-Guided Uncertainty-Aware Delayed Policy Optimization

Junqi Tu, Zejiao Liu, Fangfei Li, Yang Tang

Reinforcement learning in real world environments often suffers from severe performance degradation due to delayed feedback. Existing approaches typically mitigate performance degradation caused by observation delays by constructing augmented states or predicting the true states. However, these methods often overlook the inherent discrepancy between delayed state and true states induced by stochastic MDP. We theoretically prove the existence of such a discrepancy and show that it leads to the degradation of the optimal policy. To address this challenge, we propose Diffusion Guided Uncertainty Aware Delayed Policy Optimization (DUPO). Our method explicitly models the relationship between delayed state message and the current state using a diffusion model, and leverages the resulting discrepancy estimates to weight delayed policies. Extensive experiments on continuous robotic control tasks with multiple stochastic delays demonstrate that DUPO consistently outperforms existing methods and remains effective even under long and random delay scenarios.

View free PDFSource page

Related papers

arxivphysics.comp-phcs.AIphysics.plasm-ph2026-07-23

Cycle-Consistent and Uncertainty-Aware Neural Surrogates for Tokamak Edge Plasmas

Abdourahmane Diaw, Sebastian De Pascuale, Jae-Sun Park, Ivan Paradela Perez, Jeremy D. Lore, Stefan Dasbach

The boundary and divertor plasma govern how a tokamak exhausts power and particles, setting heat fluxes, target conditions, and the onset of detachment. Predicting these quantities is essential for operating current and future devices, but edge simulations that resolve them are t…

View free PDFSource page
arxivcs.AI2026-07-31

Don't Mix Rewards, Mix Policies: Policy Decomposition and Optimization for Multi-Reward RL

Ruiming Liang, Yi Zhong, Yizhen Yuan, Yinan Zheng, Tianyi Tan, Tianyue Wang, et al.

Modern large language models (LLMs) are expected not just to answer correctly, but to adapt their behavior to different human values and use cases. As a result, multi-reward reinforcement learning (RL) has become an increasingly important problem for LLMs, where each reward captu…

View free PDFSource page
arxivcs.AI2026-07-23

PATS: Policy-Aware Training Scaffolding for Agentic Reinforcement Learning

Yipeng Shi, Zhipeng Ma, Yue Wang, Qitai Tan, Yang Li, Peng Chen, et al.

In long-horizon LLM agent reinforcement learning, weak policies often repeat similar failures, producing uninformative rollout trajectories and limiting effective policy optimization. Existing skill-centric methods improve exploration by optimizing, filtering, or internalizing re…

View free PDFSource page
arxivcs.LGcs.AI2026-07-31

MBDiff: Multi-view Behavior-aware Diffusion Model for Probabilistic Utility Data Imputation

Rongchao Xu, Lin Jiang, Dahai Yu, Ximiao Li, Guang Wang

Utility data (e.g., electricity, water, and gas consumption), collected by ubiquitous sensors and embedded devices, often contains substantial missing values due to various factors such as device failures and data transmission issues. The data missingness can severely impact util…

View free PDFSource page