Multi-step Proximal Policy Improvement in Offline Reinforcement Learning

Offline reinforcement learning (RL) must reconcile two competing requirements: policy updates should stay near dataset-supported actions to keep value estimates reliable, yet meaningful gains often require moving beyond the behavior distribution. We develop a geometric view of offline actor updates by modeling policies as a probability manifold endowed with a chosen metric geometry. Under this lens, a broad class of offline actor objectives can be interpreted as a single proximal policy improvement step (SPI), i.e., an implicit discretization of a manifold gradient flow induced by a critic-defined energy. Building on this insight, we propose multi-step proximal policy improvement (MPI), a plug-in refinement mechanism that composes sequential re-centered proximal steps. MPI enables controlled policy improvement beyond dataset support while retaining proximal control at each refinement. The framework accommodates multiple policy geometries and admits practical instantiations for deterministic and diagonal-Gaussian policies. Experiments on D4RL benchmarks show that small numbers of MPI refinements improve strong offline baselines, including TD3+BC, ReBRAC, and IQL, on many tasks. Focused diagnostics further distinguish re-centered refinement from fixed-objective update scheduling and characterize limitations under critic error.

Recursive Value Learning for Long-Horizon Offline Goal-Conditioned RL

Scaling offline goal-conditioned reinforcement learning (GCRL) to long-horizon tasks is difficult because (1) long-range value learning depends on shorter-range estimates that may still be inaccurate, and (2) max-based value backups can amplify overestimation through repeated propagation. We propose DCRL (Divide-and-Conquer RL), which recursively decomposes each trajectory segment into a balanced binary tree and trains the values from leaves to root. Each parent is therefore updated only after its children, using an exact factorization of the observed route rather than selecting among noisy alternatives. Since this objective learns values along demonstrated routes that are not necessarily optimal, DCRL jointly propagates values across trajectories to discover shorter routes. Thanks to the balanced binary tree, DCRL reduces worst-case bootstrap depth from linear to logarithmic, and this shorter dependency structure empirically corresponds to much slower error accumulation. Across diverse goal-reaching tasks, DCRL substantially outperforms prior flat offline GCRL methods, and on the five most challenging long-horizon OGBench tasks, it improves the best prior average score from 55 to 64, surpassing all flat and hierarchical baselines.

WiSDoM: Wireless Sparse Decision Transformer with Mixture-of-Experts for Multi-Task Mobile Network Optimization

Emerging 6G wireless networks are expected to operate across diverse deployment scenarios, where variations in network topology, user mobility, traffic demand, and radio conditions challenge the scalability of conventional radio resource management (RRM). While offline reinforcement learning (RL) methods have demonstrated strong decision-making capabilities, learning a single policy that performs consistently across heterogeneous wireless environments remains difficult due to conflicting optimization objectives and limited model specialization. These challenges become particularly pronounced in coordinated multipoint (CoMP) transmission, where selecting the optimal serving-cell combination requires sequential decision-making under evolving network conditions. This paper presents the Wireless Sparse Decision Transformer with Mixture of Experts (WiSDoM), a sparse multi-task offline RL framework for adaptive multi-cell selection. WiSDoM combines Decision Transformers (DTs) with a Mixture-of-Experts (MoE) architecture that dynamically activates specialized experts according to task characteristics. This MoE mechanism improves model capacity without proportionally increasing inference cost, mitigates negative transfer, and enables expert specialization across tasks. WiSDoM is trained jointly on diverse network configurations spanning multiple base station and user equipment densities, mobility levels, and scheduler policies. Experimental results show that WiSDoM consistently outperforms heuristic methods, single-task models, and conventional multi-task DTs, improving quality of experience (QoE) by up to 55% while activating approximately one-third of the parameters of its dense counterpart during inference. Furthermore, WiSDoM exhibits strong task generalization and efficiently adapts to unseen wireless scenarios through few-shot prompting without retraining or fine-tuning.

Three Steps at a Time: Learning Representations from Action Sequences in Contrastive RL

While self-supervised approaches to reinforcement learning have achieved strong results by learning representations of states and actions, a key open question is the time scale over which actions should be modeled. Departing from the standard formulation relying on single-step actions, we extend contrastive reinforcement learning (CRL), a prototypical self-supervised method, to operate over action chunks, and find that this results in large, pervasive gains across established offline and online benchmarks: +31.7% and +93.1% across 18 and 11 environments respectively. While action-chunking-driven gains are generally explained through the ability to model non-Markovian, temporally extended policies, and to propagate unbiased multi-step returns, interestingly, we find that these arguments only partially apply to CRL. Our empirical studies suggest that, in the context of CRL, an action chunk carries more information about the goal than a single action, measurably improving the critic's representations, and rendering the algorithm significantly more effective.

PathBridger: Subgoal Bridges for Offline Goal-Conditioned Reinforcement Learning

Offline goal-conditioned reinforcement learning (GCRL) aims to learn policies for reaching diverse goals entirely from fixed trajectory data. Long-horizon offline GCRL remains challenging because sparse goal-reaching signals must be propagated over many steps, while execution errors cannot be corrected through additional environment interaction. Existing methods address these challenges by improving long-range value estimation or reducing the effective decision horizon through subgoals, options, and action chunks. In several hierarchical methods, however, a selected subgoal specifies where to go, while the intervening state-space path remains implicit in an endpoint-conditioned low-level policy. To address this interface, we propose PathBridger, a hierarchical offline GCRL method that explicitly connects subgoal selection to short-horizon execution. PathBridger constructs a state-space bridge toward the selected intermediate endpoint and decodes it into a short executable action chunk using an inverse dynamics model. Experiments across the evaluated OGBench tasks demonstrate strong aggregate performance, with particularly large gains on the multi-object Cube manipulation tasks. Code: https://github.com/SChoish/PathBridger

Learning to Allocate Incentives for Incentivized Advertising via Offline Model-Based Reinforcement Learning

Complete your ad view and grab a 5-cent bonus! In incentivized advertising, a platform promises users a bonus before observing downstream ad revenue, encouraging them to click and complete ads. It must balance the incentive promised in advance against the revenue realized afterward: insufficient incentives forfeit monetization opportunities, whereas excessive incentives reduce net profit. Because current incentives may also shape user expectations and future engagement, incentive allocation is a sequential decision problem with delayed revenue, cost sensitivity, and carryover effects. Existing work has not studied decision-making algorithms for this setting. Auto-bidding assumes available ad opportunities, while targeted promotion optimizes incentives outside the ad monetization pipeline. We formulate the problem as an MDP and develop an offline model-based RL framework for cost-controllable sequential incentive allocation. It learns a world model of user feedback and ad revenue, then performs conservative policy optimization. An independent counterfactual scorer evaluates each learned policy on held-out logs, enabling pre-launch selection without costly online exposure. Experiments on large-scale industrial data and online A/B tests show that the scorer provides a stable offline signal. The deployment path from causal inference to offline RL and then Offline-MBRL further validates the framework: MB-IQL improves per-user net profit by 7.96\% over TD3+BC, whereas reverting to plain IQL reduces it by 6.56\% (both \(p<0.0001\)).

SegBench-GC: Testing Segmentation Invariance in Multi-Step Offline Goal-Conditioned Reinforcement Learning

Offline goal-conditioned reinforcement learning (GCRL) often uses trajectory structure for future-goal sampling and multi-step targets, yet logged trajectories may be partitioned for administrative reasons that do not correspond to termination. We introduce SegBench-GC, a controlled stress test of segmentation invariance that holds transitions, source trajectories, goal sampling, optimization settings, and evaluation fixed while varying only artificial backup boundaries and whether those boundaries retain continuation value. Continuation-valid targets (CVT) provide the segmentation-consistent control: reward accumulation stops at an artificial cut, but the target bootstraps from its stored successor. In a matched-count PointMaze study with 35,000 artificial cuts, three segmentation realizations, and three optimization seeds, final 50-episode-per-task success is 50.5% uncut, 39.1% with CVT, and 19.1% when the same cuts are treated as absorbing; across segmentation realizations, naive mean success ranges from 4.8% to 31.9%. An independent published n-step baseline (n=25) from the Decoupled Q-Chunking codebase shows the same failure on Puzzle-4x5: 47.2% uncut, 58.5% CVT, and 0.27% naive across three optimization seeds. A target-level diagnostic verifies the analytic target difference to numerical precision, and learned-critic diagnostics show a large optimistic shift under naive handling while CVT remains approximately aligned with the uncut critic. CVT applies standard continuation bootstrapping rather than a new Bellman rule; the contribution is the controlled benchmark, failure isolation, and cross-learner evidence that administrative segmentation can materially change multi-step offline GCRL.

Simple Actors and Deep Critics for Scalable Reinforcement Learning

Recent progress in offline reinforcement learning (RL) has been driven by expressive generative actors such as diffusion and flow-matching policies, which capture multimodal behavior in offline datasets. However, these actors require multiple denoising or integration steps per action and thus incur substantial overhead at every decision in deployment. In this work, we revisit where capacity should be invested in an offline actor--critic method. Since the critic is used only during training and is discarded at deployment while the actor runs at every decision step, allocating capacity to the critic rather than the actor is more favorable for inference-time efficiency. However, scaling MLP critics in offline RL is known to introduce several distinct instabilities that have, in practice, kept critics shallow. We identify three distinct failure modes that arise when critics are deepened in offline RL---optimization, bootstrap-noise amplification, and value-range drift---and address each with a corresponding ingredient: a residual MLP backbone, n-step bootstrap targets, and a categorical cross-entropy loss. Combining these ingredients with a lightweight deterministic actor, we propose LAC (Light Actor, deep Critic). On OGBench, LAC matches the strongest diffusion- and flow-matching baselines while achieving up to 4x lower inference latency, comparable to one-step distilled policies without distillation. Its critic recipe also transfers across actor parametrizations.

Towards Scaling Reinforcement Learning to Massive Populations: Learning Mean-Field Representations

Modern multi-agent systems are increasingly deployed at scale over large populations of agents in settings such as ad-auctions, traffic routing, and recommendation systems. The dominant approach in such settings is to optimize each agent's policy independently, treating the other agents as part of a fixed single-agent environment rather than modeling the population dynamics. In many large-population systems, the dynamics depend on an aggregate summary of the population rather than the identity of any individual. Mean-field RL exploits such structure, providing a principled framework that models each agent's environment as an explicit function of the population distribution. However, in large state-action spaces or high-dimensional control problems, modeling the population distribution is itself intractable. How can we design a scalable framework for high-dimensional control problems with large populations? This work explores this question from the perspective of representation learning. We introduce a mean-field RL framework in which the rewards and transition dynamics depend on the population only through an unknown low-dimensional aggregate statistic. We then study this framework in the offline setting and design a provable approach that learns a near-optimal policy by learning a low-dimensional representation. Motivated by real-life supply-chain optimization problems, we design a one-step routing game to test the hypothesis that learning a low-dimensional population representation improves reward prediction and Nash gap estimation relative to baselines that don't exploit this structure. We show that under a fixed neural-network parameter count and optimization budget, learning a low-dimensional population representation improves reward prediction and the equilibrium quality of the resulting policies.

AERIS: Offline Policy Improvement for Multi-UAV Integrated Sensing and Communication

Unmanned aerial vehicle (UAV)-enabled integrated sensing and communication (ISAC) is a promising 6G paradigm, but dynamic multi-UAV ISAC control must jointly balance communication quality, sensing reliability, and flight safety under stochastic mobility. Existing optimization methods often require repeated global non-convex solving, while online reinforcement learning (RL) depends on risky trial-and-error flights that may cause sensing loss or collision-risk events. This paper proposes AERIS, an offline policy improvement framework for multi-UAV ISAC. AERIS learns from fixed flight logs under centralized training and decentralized execution, so each UAV acts from local histories while training uses logged global information to assess team-level effects. We further design STAR-CRDT, an offline multi-agent RL algorithm that performs support-aware local action rectification and distills only trusted improvements into the decentralized actor. We prove an offline-support policy improvement guarantee. Experiments show that STAR-CRDT improves the main ISAC objective return by 29.3% over the strongest baseline. It further improves communication sum rate, sensing pass rate, and sensing margin by 3.4%, 4.8%, and 69.1%, while reducing collision-risk events by 54.2%. On unseen real-road maps built from OpenStreetMap data, STAR-CRDT still obtains the best return.

Bayesian Flow Networks for Offline Trajectory Planning

Offline reinforcement learning (RL) leverages static datasets to learn decision policies without real-time environment interaction. While recent sequence-modeling approaches rely on continuous diffusion models for trajectory synthesis, applying these methods to discrete planning tasks requires a categorical formulation rather than the standard Gaussian construction. We present BFN-RL, a unified generative modeling framework for offline RL based on Bayesian Flow Networks (BFNs). By iteratively evolving distribution parameters rather than noisy data instances, BFN-RL natively models both discrete and continuous trajectory spaces within a single probabilistic formulation. The categorical planner generates future state sequences, and a learned inverse-dynamics model converts consecutive generated states into actions. Evaluations in discrete planning and continuous control show that BFN-RL can generate effective trajectories across both categorical and continuous state spaces. Our results establish BFNs as a versatile generative foundation for offline trajectory planning across data modalities.

Bellman Calibration for Marginalized Importance Weighting in Offline Reinforcement Learning

Marginalized importance weighting evaluates a target policy by reweighting offline state-action samples with its discounted occupancy ratio, characterized by an adjoint Bellman equation. Existing minimax, primal-dual, and fitted fixed-point estimators can leave residual occupancy-balance violations because of function-class approximation, regularization, or incomplete optimization. These violations are difficult to diagnose and reduce because the objectives generally lack a direct supervised validation loss for hyperparameter tuning, model selection, and early stopping. We introduce isotonic Bellman calibration, a one-dimensional, model-agnostic post-processing method that reduces these violations while preserving the ranking information in any initial occupancy-ratio estimate. The method corrects the estimate's scale and shape by applying fitted occupancy-ratio evaluation (FORE) over a one-dimensional class of nondecreasing transformations. We characterize Bellman calibration as a conditional fixed-point property equivalent to occupancy-balance against every test function of the calibrated ratio. More generally, we derive a calibration-refinement bound showing that any fitted ratio with small calibration error performs nearly as well as the best post-processing based on its fitted values. For isotonic Bellman calibration, we establish finite-sample calibration guarantees and a KL oracle inequality relative to the best monotone transformation of the initial estimate. Consequently, isotonic Bellman calibration achieves small calibration error and KL risk within statistical error of the best monotone correction, with guarantees for downstream target-occupancy functionals, including policy-value estimation.

CoDrift: Compositional Drifting for Offline Reinforcement Learning

Offline reinforcement learning is intrinsically multi-objective: a policy must remain compatible with the behavioral support of a fixed dataset while preferentially selecting high-value actions. We recast these objectives in a common form by viewing each as an action-space motion field that specifies how generated actions should move. This perspective enables heterogeneous learning objectives to be combined directly through field composition. Inspired by drifting models, we propose CoDrift, a compositional framework for one-step generative policy learning. CoDrift combines three objective-level fields into a unified policy field. The conditional field preserves state-dependent behavioral structure, while the marginal field pools actions across states to provide a more stable generative signal in the single-positive-sample regime of continuous-control offline RL. The value field moves generated actions toward higher-value regions. The composed field is absorbed into a stochastic generator that produces an action with a single forward pass at deployment. We evaluate CoDrift on 73 tasks from OGBench and D4RL in both offline and offline-to-online settings. CoDrift compares favorably with state-of-the-art methods and achieves the best average rank in both settings.

Learning from the Test: Self-Referential Differential Testing for Deep RL Agents

Deep Reinforcement Learning (DRL) has achieved significant success in complex decision-making problems. As DRL systems are increasingly deployed in real-world applications, ensuring their quality and reliability is paramount. Current works primarily focus on detecting safety-critical failures, often neglecting policy optimality, which can lead to reduced efficiency, user distrust, and economic losses. This oversight, compounded by the inherent "testing oracle problem" for optimality, leaves a significant gap in comprehensively evaluating DRL systems. To address this gap, we propose Delta (Differential Testing for DRL Agents), a novel and comprehensive framework that automatically identifies both safety-critical and optimality bugs in DRL agents. Delta employs a two-phase approach: (1) Safety Testing, where the Agent Under Test (AUT) is evaluated for catastrophic failures while collecting data from its decision-making policy, and (2) Optimality Testing, where this collected data from the prior phase is used to train a challenger agent via Offline Reinforcement Learning. Differential testing is then performed by comparing the challenger agent against the AUT; instances where the challenger achieves higher cumulative rewards indicate optimality issues in the AUT. We demonstrate Delta's effectiveness across five environments. We investigate the effectiveness of three offline RL algorithms (BC, BCQ, and CQL) in generating challenger agents. Experimental results demonstrate that safety testing datasets are valuable for training competent DRL agents. Challenger agents trained with BCQ proved most effective for identifying optimality issues within the framework of Delta. Across the five environments, Delta uncovered an average of 2,518 optimality issues, outperforming the baseline methods by 50.2%.

Decoupling Policy Extraction for Offline Reinforcement Learning

Offline RL methods commonly jointly train the actor and critic, where the critic is used to guide the actor toward higher-value actions. This coupled learning process is well motivated in online RL, where an improved actor collects new data that can further update the actor and the critic. However, training data remains fixed in offline RL, making actor-side policy improvement unable to generate new data to validate or correct the critic. Moreover, retaining this coupled paradigm leads to two related challenges. Firstly, actor updates can drift toward high-valued but potentially out-of-distribution (OOD) actions and amplify critic overestimation. Secondly, conservative value estimation or behavior-cloning regularization creates a difficult trade-off between suppressing OOD actions and selecting high-value actions within the data-supported region. Motivated by this observation, we revisit the conventional offline RL paradigm and propose decoupling policy improvement from actor training. Specifically, we train the actor solely to model the behavior distribution and perform policy improvement at inference time by reranking multiple actor-generated proposals with a separately learned critic. We refer to this paradigm as the decoupled policy extraction paradigm. Under such paradigm, the actor provides behavior-supported action candidates, while the critic performs value-based selection within this candidate set. Extensive experiments show that the decoupled policy extraction paradigm outperforms both behavior cloning and jointly learned offline RL methods, while remaining effective even with a naive Q-learning critic.

RoMAN-Flow: Taming Autoregressive Normalizing Flows for Offline Reinforcement Learning in Robotic Manipulation

Offline reinforcement learning improves robotic policies using previously collected data without further environment interaction. Yet prevalent diffusion- and flow-matching robot policies lack tractable likelihoods, limiting their use in likelihood-based offline RL post-training. AR-NFs offer both expressive action modeling and exact likelihood evaluation, but their sequential sampling incurs substantial sampling overhead during policy optimization and deployment. We present RoMAN-Flow (Robotic Manipulation with Autoregressive Normalizing Flows), an offline reinforcement learning framework that makes AR-NF policies practical for robotic manipulation by addressing this sampling bottleneck in both stages. During policy optimization, RoMAN-Flow employs a sampling-free, advantage-weighted likelihood objective that assigns higher likelihood to high-advantage actions from the offline dataset without sampling from the autoregressive policy. For efficient deployment, it distills the optimized autoregressive policy into a one-step action generator, enabling low-latency action prediction. Experiments across multiple simulated manipulation benchmarks and real-world robotic platforms demonstrate that RoMAN-Flow achieves competitive policy performance while substantially reducing inference latency. Code is available at https://github.com/konnyaku28/RoMAN-Flow.

Learning Hierarchical Skill Policies with Offline Quality-Diversity Reinforcement Learning

Recent studies investigate how to leverage pre-collected datasets to improve the policy performance and sample efficiency of RL. One promising approach to achieve this goal is to employ a two-stage strategy: In the first stage, diverse skills are extracted as a low-level policy from a given dataset, and a high-level policy is trained to solve a specific task in the second stage. Typically, extraction of the low-level policy is performed based on unsupervised learning such as trajectory VAE. However, a limitation of this approach is that the quality of the low-level policy highly depends on the quality of the dataset. To address this issue, we introduce QDOS (Quality-Diversity Offline Skill learning), a unified pipeline for robust offline-to-online learning. Our approach incorporates an Advantage-Weighted Quality-Diversity pretraining objective, which weights the skill extraction and diversity objectives by the estimated advantage of each trajectory segment. This approach allows the model to extract diverse and high-value skills. By providing robust and task-relevant skill representations, QDOS significantly improves the quality of the embedded skill space used by the low-level policy. We further integrate this with a dual dataset reuse strategy, where offline data is used both for skill pretraining and for populating the online replay buffer via pseudo-labeling. Experiments demonstrate that QDOS significantly outperforms strong baselines in structured manipulation tasks and unstructured locomotion tasks, confirming its ability to accelerate exploration and improve final returns in challenging sparse-reward domains.

Offline Reinforcement Learning for Hemodynamic Management of Sepsis in the ICU: a MIMIC-IV Study with Dual Off-Policy Evaluation

The dosing of intravenous fluids and vasopressors in sepsis is a sequential decision made under uncertainty and guided largely by clinical judgment, which makes it a natural target for reinforcement learning from historical care. Because a learned policy cannot be trialed on patients, its value must be estimated off-policy, and such estimates can be fragile and optimistic. This work advances the reliable evaluation of sepsis treatment policies by combining off-policy estimation, reliability diagnostics, and clinician-agreement analyses in a transparent validation framework. We modeled fluid and vasopressor dosing on a cohort of 36,872 septic ICU stays drawn from the MIMIC-IV critical-care database, as a discretized Markov decision process with 1,000 states and 25 actions, defined by a five-by-five grid of fluid and vasopressor levels and solved by policy iteration. The clinicians' behavior policy was estimated with a random forest, which mitigated the collapse of the Effective Sample Size (ESS 50.1 against 4.0 with smoothed counts) that otherwise destabilizes the importance-sampling estimate. The learned policy was evaluated with two estimators, weighted importance sampling (WIS) and fitted Q evaluation (FQE), with the ESS and clinician agreement as reliability checks. An empirical variable selection found that the composition of the state matters more than its size. Both estimators place the learned policy above the clinicians' return (WIS 50.8 and FQE 46.8 against 38.2, ESS 50.1), yet it departs only modestly from observed practice (total variation 0.18), favoring less intravenous fluid. These retrospective single-center off-policy results support the learned policy as a clinically plausible refinement of observed practice and motivate its further evaluation as a discordance-based clinical decision-support approach.

Offline Deep Q* Estimation with Diffusion Models

In offline RL, estimating the optimal action-value function $Q^*$ can be formulated as solving the optimal Bellman equation based solely on offline observations. A fundamental challenge is that the reward function and transition kernel are unknown, so the optimal Bellman operator is not directly observable from data. To address this issue, we propose a novel framework that decouples operator estimation from value function learning. In this approach, we first formulate conditional diffusion models to estimate the reward law and transition kernel, which induces a data-driven approximation of the optimal Bellman operator. We then plug these estimators into the Bellman equation and obtain a deep estimator of $Q^*$ by minimizing the empirical Bellman residual over a neural network function class. Theoretically, we first establish sharp nonasymptotic convergence rates for learning the optimal Bellman operator through an end-to-end analysis of conditional diffusion estimation in total variation distance. We then establish the oracle value-stage rate $\widetilde{\mathcal O}\bigl(n^{-\frac{2β}{d_x+d_a+2β}}\bigr)$ for the excess Bellman residual risk. Finally, under a concentrability condition, we translate this residual bound into an $L^2$ convergence rate of $\widetilde{\mathcal O}\bigl(n^{-\fracβ{d_x+d_a+2β}}\bigr)$ for the resulting deep estimator of $Q^*$, where $d_x$ and $d_a$ denote the dimensions of the state and action spaces, respectively, and $β$ denotes the Hölder smoothness index of $Q^*$. Importantly, our theoretical analysis does not rely on completeness assumptions commonly used in deep RL theory. Extensive numerical experiments demonstrate the effectiveness of the proposed method and its strong empirical performance.

Redistribution-based Cost Inference Improves Sparse Safe Offline RL

Safe offline RL typically assumes access to dense per-step cost annotations, but in practice supervisors provide only trajectory-level stop-feedback: a binary signal at the first unsafe transition, with no per-step attribution. We frame this as a temporal credit assignment problem and propose the Redistribution-based Cost Inference (RCI) framework, which converts sparse stop-feedback into dense per-step costs via return decomposition, then trains a constrained offline policy on the augmented dataset. We show that return-equivalent redistribution preserves the feasible policy set and the optimal Lagrangian in a CMDP, establishing that the transformation is lossless in theory while yielding better-conditioned cost critic learning in practice. Experiments on highway driving and robotic manipulation demonstrate substantially lower violation rates than sparse and classifier-based baselines, with robustness to heterogeneous dataset compositions and label noise.

Unmasking Toxic Mimicry in Medical Offline Reinforcement Learning for ICU Sepsis Management via Counterfactual Clinical Audits

Offline reinforcement learning (RL) offers considerable promise for optimizing ICU treatment decisions, yet standard evaluation metrics Mean Squared Error (MSE) and Fitted Q-Evaluation (FQE) assess only behavioral imitation and cannot detect Toxic Mimicry, a failure mode in which agents replicate harmful patterns such as treatment withdrawal during comfort-care transitions. Using the MIMIC-III database, we propose the Counterfactual Clinical Audit (CCA) framework, which stress-tests RL agents through physiological perturbations anchored in Surviving Sepsis Campaign (SSC) guidelines. We audit a Medical Decision Transformer (MedDT) and a Historical Causal Transformer (HCT-RL), the latter employing Causal Action Shielding, propensity-based importance weighting, and Conservative Q-Learning. CCA reveals that MedDT paradoxically reduces vasopressor dosage as lactate escalates, contradicting resuscitation guidelines, while HCT-RL maintains physiologically consistent responses. These findings expose a systemic misalignment between statistical fit and clinical safety, supporting counterfactual audits as a necessary evaluation standard for medical RL.

Dynamics Models for Offline Hyperparameter Selection in Real-World RL

A key obstacle to deploying reinforcement learning in real-world systems is hyperparameter selection, particularly when simulators are unavailable and online experimentation is costly. Prior work has proposed calibration models trained on offline data to approximate environment dynamics and enable offline hyperparameter selection, but these methods have so far been evaluated only in simple simulated settings. In this paper, we present the first application of calibration models in a real-world industrial setting: a municipal water treatment plant. We evaluate several calibration model approaches, including a k-nearest neighbors model with a Laplacian distance metric, on high-dimensional, non-stationary sensor data for nexting prediction tasks. Our results show that these models can generate realistic long-horizon rollouts and recover meaningful hyperparameter sensitivity trends. We further examine how calibration models scale to year-long datasets, how they support the selection of fine-tuning learning rates for pre-trained agents, and how robust they are under distribution shift. Overall, our findings provide a proof of concept for using offline dynamics models to support RL deployment in real-world environments, while highlighting important practical challenges for future work.

CODS: Iterative Bellman-Residual Data Selection for Reusable Offline Reinforcement Learning

Offline reinforcement learning repeatedly trains policies from a fixed transition pool, making redundant data costly across seeds and hyperparameters, while naive subsampling can remove rare transitions needed for long-horizon credit assignment. We introduce CODS, a critic-guided selector that alternates between fitting an algorithm-matched critic and acquiring high-residual transitions before freezing a reusable subset. Unlike prioritized replay, CODS produces a static artifact; unlike one-shot residual selection, it refreshes scores as the critic changes. At a 10\% budget, CODS retains 96.6\% of eligible-pool performance across 20 valid D4RL task--algorithm cells. It exceeds ReDOR and OPER on 19/20 cells and every other subset baseline on 20/20; all six subset advantages remain significant under predeclared hierarchical inference with Holm correction. Holding total selector updates fixed, five acquisition rounds improve four representative cells by 11.23 points over one round and saturate thereafter. Equal-pass and equal-hour evaluations clarify that reuse, rather than a single-run speedup, creates the compute advantage. Mechanism and corruption interventions expose both useful sparse-reward enrichment and sensitivity to outliers. Finally, a whole-trace extension retains 95.4\% of pooled ALFWorld success and 96.5\% of pooled GSM8K exact match. CODS is therefore a reusable selection procedure, not a formal coreset guarantee.

Convex-Hull-Neighborhood Smooth Dual Generalization: Controlling Local Correction Propagation in Offline RL

Offline reinforcement learning (offline RL) can benefit from nearby out-of-distribution (OOD) actions, but estimation errors at these actions may be amplified by bootstrapping. Existing regularization and local-generalization methods control either the admissible OOD region or the influence of generalized targets, often through separate mechanisms. We propose Convex Hull Neighborhood Smooth Dual Generalization (CSDG), which expresses the Bellman backup as an in-sample value target plus a CHN-local correction. This formulation makes the generalized contribution explicit and separates it from the in-sample reference path. The correction is obtained by smoothing in-sample-oriented and OOD-oriented candidates sampled at different perturbation radii. A mixture coefficient lambda scales its contribution to each backup, while the recursive discount remains gamma. Under boundedness and fixed perturbation kernels, we derive an exact one-step correction identity, a time-varying iterate bound, and a fixed-point bound that depends only on the branch discrepancy at the fixed point. We further characterize the implicit policies induced by the idealized operators and give a conditional non-degradation criterion. The practical algorithm approximates these quantities using asymmetric bounded noise and expectile regression, without exact support classification or an additional pessimistic OOD penalty. Experiments on Gym-MuJoCo and AntMaze show strong aggregate performance and stable value estimation. Code is available at: https://github.com/YOUNG-fnxm/CSDG

Diffusion Policy with Behavioral Advantage Correction for Offline Reinforcement Learning

In offline reinforcement learning (RL), the distribution shift between behavioral data and the learned policy can lead to erroneous \emph{Q}-value estimation, thereby misguiding the direction of policy optimization. To address this issue, we develop a behavioral advantage corrected policy evaluation (BAC-PE) approach, which utilizes the \emph{Q}-function of the behavior policy to correct the learned policy's \emph{Q}-function, thus mitigating pessimistic conservatism and overestimation bias. Furthermore, the convergence of BAC-PE is analyzed theoretically, and an upper bound on the difference between the learned \emph{Q}-function and the true \emph{Q}-function is derived. To alleviate distribution shift, this work employs diffusion models to represent both the behavior policy and the learned policy, performing distribution matching for accurate policy regularization. Additionally, \emph{Q}-value guidance is incorporated into the training process to achieve effective policy improvement. By combining BAC-PE with diffusion policy modeling, we propose the diffusion policy with behavioral advantage correction (DPBAC) algorithm. Compared to existing offline methods, DPBAC demonstrates stronger policy representation capabilities and effectively mitigates the bias in \emph{Q}-value estimation. Experimental results on multiple domains of D4RL tasks show that DPBAC achieves superior performance, with notable advantages over state-of-the-art (SOTA) algorithms.

ReBRAC-v2: The Return of the King

Recent offline reinforcement learning methods increasingly rely on expressive generative policies and specialized value-guidance mechanisms. We ask whether comparable progress can instead come from systematically modernizing a conventional behavior-regularized actor-critic while preserving its algorithmic simplicity. We introduce ReBRAC-v2, which directly trains an exact-likelihood normalizing flow as the RL actor, combines likelihood, MSE, and MAE behavior regularization, and integrates a classification-based residual critic, staged optimization, and multi-sample test-time action selection. Rather than tuning this recipe separately for every task, we develop a single shared configuration via roughly 600 Bayesian proposals on six challenging OGBench tasks, freeze all structural and optimization choices, and adapt only two behavior-regularization coefficients over a 16-point grid. Across ten common state-based OGBench categories, ReBRAC-v2 averages 74.8 compared to 52.3 for the next-best aggregate result and ranks first in eight categories. The same recipe, without structural changes, obtains the strongest averages in our comparisons on D4RL AntMaze (90.2) and Adroit (33.6). Fixed-recipe ablations show the largest sensitivity to the selected mixed cloning objective, staged training, sufficient flow capacity, and multi-sample inference, while showing that several smaller choices depend on the values of other hyperparameters. These results show that disciplined, transferable engineering can achieve state-of-the-art aggregate performance without abandoning a minimalist offline RL foundation.

RedFlow: Redirect Failure into Action-Level Corrections for Flow-matching VLA Policy

Flow-matching Vision-Language-Action (VLA) policies have shown strong potential for robotic manipulation but often suffer from compounding errors caused by distribution shifts during deployment. While offline reinforcement learning (RL) provides a practical way to improve deployed policies using rollout data, existing methods either ignore failure data or exploit it only at the trajectory level, resulting in low learning efficiency and persistent errors. We propose **RedFlow**, a fine-grained offline RL framework that redirects failure experiences into action-level corrective supervision for flow-matching VLA policies. RedFlow consists of two key components: (1) a **Context-Aware Corrective Matching** mechanism that identifies failure-inducing actions and retrieves successful alternatives from similar contexts as corrective targets, and (2) an **Adaptive Redirection Objective** that jointly reinforces successful actions, suppresses undesirable ones, and redirects recoverable failures toward corrective targets. By converting both successful and failed experiences into dense supervision, RedFlow enables robust recovery learning from mixed-quality data. Experiments on the LIBERO benchmark and three real-world manipulation tasks show that RedFlow consistently outperforms state-of-the-art offline RL baselines, improving the real-world success rate from 56.7% to 74.7%. It also matches strong on-policy methods (PPO, GRPO, and DDPO) while requiring roughly an order of magnitude fewer training samples.

Learning Implicit Causal World Models from Multi-Agent Demonstrations

In model-based reinforcement learning, world models exist as internal simulators, but their training often conflates statistical correlations with causal mechanisms. This problem is exacerbated in multi-agent systems where physical transitions are intertwined with strategic agent intents, causing world models to fail under distribution shift. We introduce Implicit Causal World Models to recover environmental dynamics from offline demonstrations without requiring pre-defined causal graphs. By incorporating policy variance, we render world models discoverable via the sequential backdoor condition. Evaluations across coordination tasks (Two-Door, Navigation, and Giveway) demonstrate that these models provide interpretable causal representations under both full and partial observability, with model accuracy scaling directly with interventional strength.

Learning from the Unseen: Offline Reinforcement Learning with Hidden Actions

Standard offline reinforcement learning (RL) algorithms typically assume that the actions in the dataset are observed without error. However, in many real-world applications, the true actions are unobserved and only noisy proxies are available, causing existing RL methods to yield biased and potentially misleading conclusions. We study off-policy evaluation in infinite-horizon discounted Markov decision processes with hidden actions. By leveraging the next-state variable as a natural proxy for the unobserved action, we establish identification of the policy value and propose an influence-function-based estimator called LURE (Learning from the Unseen: Robust Estimator). LURE is multiply robust, remaining consistent under several combinations of correctly specified nuisance components, and is asymptotically normal, enabling valid statistical inference. To our knowledge, this is the first work to address offline RL with hidden actions. We demonstrate LURE's effectiveness through simulations and a sepsis management application using the MIMIC-III database.

TOUR: A Trajectory-Level Unlearning Benchmark for Offline Reinforcement Learning

Offline Reinforcement Learning (RL) agents are trained on fixed behavioral trajectories, which makes trajectory-level deletion important when selected data must be removed after training. Evaluating such deletion is difficult because a lower membership score can reflect trajectory removal, residual memorization visible to another attack, or policy collapse that destroys useful behavior. We introduce Trajectory-level memOrization and Unlearning in offline RL (TOUR), a benchmark that combines trajectory-level partitioning, matched non-member controls, retraining references, retained-performance anchors, and multi-attack privacy auditing. Across D4RL locomotion experiments and an exploratory AntMaze extension, TOUR shows that common deletion baselines have environment-dependent privacy-utility behavior. Retraining and fine-tuning often provide stronger retained-utility references than uniform GA+Refit, while TrajDeleter remains a useful comparator but is not uniformly stronger under the same audit. Reference-model, threshold, deviation, equivalence, action-error, representation-based, and query-limited attacks further show that a single likelihood-based membership score can overstate deletion quality. In the evaluated settings, conclusions about offline RL unlearning are therefore not stable under single-score auditing. They depend on matched non-member construction, retraining-relative calibration, attack family, retained utility, and explicit scope for diagnostic architecture or component-level evidence.