Soft Actor-Critic

Reinforcement learning has become a central component of large language model (LLM) post-training, yet token-level credit lacks a generally accepted mathematical definition, leaving its relationship to commonly used training signals unclear. We formulate three regularity conditions, namely Completeness, Prefix Consistency, and Neutrality, and prove that they uniquely determine token-level credit. This characterization provides a unified basis for explaining phenomena across existing algorithms and guides the development of an improved actor-critic training procedure. Through this lens, an ideal teacher in On-Policy Distillation (OPD) acts as an implicit critic, yielding an expected policy gradient proportional to that induced by token-level credit. Response-level REINFORCE Leave-One-Out (RLOO) signals match the expected policy-gradient contribution of token-level credit despite their coarser granularity. We further establish approximate credit sparsity under bounded outcome rewards and show how intermediate critic errors in Generalized Advantage Estimation (GAE) can become comparable to the underlying credit. These motivate Policy Aligned Critic Training (PACT), which adopts an Actor-then-Critic update order to apply importance sampling correction to critic training and better align the critic with the updated policy. In agentic mathematical reasoning, PACT achieves 72.87% average accuracy across four benchmarks, outperforming GRPO and PPO by 8.80 and 13.16 percentage points, respectively. On SWE-bench Verified, PACT achieves a pass rate of 67.4%, outperforming PPO, GRPO, and SAO by 2.4, 2.0, and 3.8 percentage points, respectively.

Reliable temporal-difference targets are central to off-policy actor-critic learning. Direct value improvement refines the next-state target with alternative actions, but the reliability of this refinement depends on how candidate actions are ranked, reviewed, and weighted. Noisy rankings may force premature candidate commitment, reusing selection scores may bias target valuation, and fixed enhancement weights may amplify weak evidence. To address these risks, we develop Conservative Adaptive Ranking and Screening (CARS), which retains an ordered candidate prefix within a preset budget and narrows it only when the observed boundary gap exceeds a disagreement-scaled uncertainty radius. Selector-Evaluator Value Assessment (SEVA) uses selector critics to order candidates and a separately parameterized evaluator critic to review the selected value, then caps the reviewed value at the selector reference. Dynamic Adaptive Risk-aware Enhancement (DARE) then regulates each residual correction using candidate reliability, the gap between selector and evaluator signals, and a finite stage factor. Together, CARS, SEVA, and DARE form CARE-VI, an evidence-regulated target construction framework that preserves the backbone interfaces for critic regression and actor updates. The analysis bounds the CARS boundary error, the SEVA selected-value overestimation, and the one-sided deviation of the DARE residual displacement from its population counterpart, and establishes fixed-policy recovery after the finite-stage perturbation ends. Experiments with SAC, TD3, and TD7 on four MuJoCo tasks show that CARE-VI achieves the highest mean return in all twelve settings. Grouped ablations and scalar diagnostics support the roles of the three components in improving target reliability.

Actor-critic architecture has been widely used in continuous robot control. However, they rely on learning a value network, introducing additional computational overhead during training. Moreover, policy learning may also be affected by the approximation error of value estimation. Critic-free group relative policy optimization methods provide a simpler training approach by removing the need for a critic. However, they fail to learn long-term action outcomes when directly applying immediate rewards to policy optimization in dense-reward environments. To address these problems, we propose Group Relative Return Policy Optimization (GR2PO), a critic-free reinforcement learning framework for continuous robot control. GR2PO estimates the discounted returns from the parallelly collected trajectories, performs group normalization at each rollout time index, and uses relative advantages and clipped targets to update the policy. To evaluate the effectiveness of the proposed framework, we instantiate it on robot control simulation environments and deploy the model to a real-world edge device. The results show that GR2PO significantly outperforms critic-free baselines that use immediate rewards and performs competitively against state-of-the-art actor-critic methods. Furthermore, GR2PO demonstrates competitive training efficiency. Inference tests on NVIDIA Jetson TX2 demonstrate the feasibility of deploying the learned policies on edge platforms. Further ablation experiments analyze the effects of parallel group size, return estimation methods, and target clipping ratio on learning performance. To support follow-up research, we will make the complete code publicly available after the paper is accepted, including the framework implementation, experimental configuration, and training and evaluation scripts.

In reinforcement learning for large language models, Proximal Policy Optimization (PPO) commonly uses a critic to estimate state values and reduce the variance of policy updates. However, we uncover a systematic failure mode in PPO critics, which we call Value Flattening: state values, estimated from multiple Monte Carlo continuations, change sharply across intermediate states while critic predictions remain comparatively flat. We further observe this phenomenon in a controlled FrozenLake environment and find that it becomes more pronounced as the state space grows. Our theoretical and empirical analyses relate Value Flattening to an implicit variance penalty in the critic loss and redundant updates from temporally correlated states with similar gradients. Motivated by these findings, we introduce SParse Proximal Policy Optimization (SP$^3$O), which applies the value loss to only a few well-separated states in each response to mitigate both effects. Experiments on Qwen3-Base show that SP$^3$O with only three states supervised per response can mitigate Value Flattening and consistently improve the learned policy across model sizes and evaluation suites. Together, our results identify Value Flattening as an important yet overlooked failure mode of critic learning in standard PPO and show that a simple sparse supervision strategy can mitigate it.

Aerial grasping is a remarkable capability exhibited by predatory birds, allowing them to capture prey through highly coordinated maneuvers in flight. Inspired by this capability, researchers have developed various formulations to reproduce such maneuvers through trajectory optimization. However, two limitations remain in practice. First, the resulting optimization problem is highly nonconvex and sensitive to initialization, making high-quality solutions difficult to obtain under a limited computational budget. Second, prescribed numerical objectives are human-designed abstractions that describe successful grasping through a limited set of mathematically tractable quantities and may not fully capture what determines task success. We investigate how learning can address these limitations within an analytical planner. Accordingly, a trajectory prior is first learned from optimized motions and then evolved through a CEM-based process that evaluates sampled initializations with the deployed optimizer and retains favorable ones as new supervision. An Execution-Aware Critic learns from contact, lift, and completion outcomes to assess whether the optimized trajectories are likely to succeed in physical execution. Its frozen energy can further serve as a differentiable grasping cost, allowing execution data to directly shape trajectory generation. Simulation and real-world experiments demonstrate improved optimization reliability, trajectory consistency, and grasping performance.

Reinforcement learning is now the standard way to train large language model agents on long-horizon tasks, where dozens of interdependent actions precede a single sparse reward. Critic-free, group-relative methods such as GRPO suit this regime, but they broadcast one trajectory-level scalar to every step and cannot say which decision drove the outcome. GiGPO recovers a step-level signal by grouping time steps that share an anchor state, yet it merges the step- and episode-level estimates under one fixed weight, spending the same resolution on a pivotal branching decision as on a routine, near-deterministic transition. We argue that the right resolution is state-dependent, and propose GACA, a critic-free estimator whose granularity follows an uncertainty-based criticality proxy. GACA scores every step by the negative log-likelihood its own rollout already records, then blends the two advantages with a per-step weight that grows with that score, so the gradient places more weight on the fine-grained signal at above-average NLL and on the episode-level signal below it. We derive an exact risk decomposition for the implemented mixture and show that sufficiently small modulation improves on fixed mixing under positive directional alignment. A separate conditional result bounds local action-value variation using expected NLL, while an error-projection analysis characterizes when mixing adds value beyond scalar uncertainty reweighting. On ALFWorld and WebShop, GACA improves task success over GRPO and GiGPO at both 1.5B and 7B scales.

Tree-structured rollouts give critic-free reinforcement learning with verifiable rewards (RLVR) step-level credit: fork a chain at an intermediate point, and sibling outcome differences estimate step value. Each fork adds sampling cost, so realistic budgets typically allow only a few forks per chain. A fork placed where the outcome is already largely settled yields siblings that mostly agree and provide almost no credit signal; hence, for a given tree size, where forks are placed largely determines how much step-level RL can gain. Most existing mainstream methods place forks by structure, such as fixed lengths, midpoints, and delimiters, or by next-token entropy. We formalize fork placement as locating the \emph{pivots} of the chain's value curve, where the expected outcome turns. We propose \emph{belief-shift branching}: read the model's answer belief at candidate boundaries and fork just before the step where consecutive beliefs diverge most. Three instantiations, none needing step-level supervision, span access levels: a black-box probe, a logit-lens depth profile, and a learned activation direction, which is fit offline and therefore used only in the validation before RL training. The signal only \emph{places} forks, and the probe costs about $1\%$ of step compute on mathematics and under $5\%$ on code when it runs inside the rollout engine. In that validation, against Monte-Carlo value curves, a belief-shift signal ranks first in each of the eight model$\times$benchmark panels, ahead of entropy, structural, and LLM-judge baselines. In RL across three model families and two domains, belief-shift forking leads every mathematics aggregate, on OLMo-3-7B by $+2.6$ aggregate and $+2.9$ on AIME 2026 over the strongest baseline, and sweeps every OLMo code column, by $+6.5$ on LiveCodeBench-medium.

Agent usage is shifting toward long-horizon tasks such as coding and scientific discovery, among which terminal tasks are especially important. We introduce T1, a Mixture-of-Experts model of 122B total trained with reinforcement learning, operating a real shell in a cloud sandbox for up to 300+ tool-call turns per task, rewarded by executing each task's own verifier. We provide a comprehensive recipe: First, an aggressively warm-started to stabilize actor-critic training, with a dense process reward scoring trajectories by the absolute number of passing verifiers. Second, stable optimization through TITO construction, training on the exact sampled token identifiers with drift repair at turn boundaries, and rollout routing replay, recording the sampler's per-token expert choices at every MoE layer and replaying them during training. Third, fully out-of-distribution training corpus: isolated seeds and synthesized tasks disjoint from Terminal-Bench 2.1 ensures gains reflect genuine capability transfer over benchmark overfitting. Together, TITO and R3 cut the training-to-inference log-probability difference from 0.021 to 0.013, with exactly aligned zero token drift in the loss region. On Terminal-Bench 2.1, our post-train pipeline raises initial base model from 43.8% to T1 with 64.0% resolved. On Long-Horizon Terminal Bench, T1 reaches 27.9% and surpasses GPT-5.4 and GLM-5.1.

Soft Actor-Critic (SAC) represents a continuous policy as an unbounded Gaussian that is squashed by tanh. The Jacobian of that map is $\partial a/\partial u = 1-a^2$, which vanishes as $|a|\to 1$. A natural concern is that this throttle starves the actor of critic signal exactly where extreme actions (full brake, full throttle) are optimal. We test a minimal intervention that restores the missing signal: one extra term in the actor loss whose gradient on the pre-tanh mean is the detached action-gradient of $Q$, with no gain parameter. On a minimum-time double integrator whose optimum is bang-bang at the action bounds, vanilla SAC already reaches near-optimal return ($-31.6$ vs. a calibrated optimum of $-30.3$) across ten paired seeds. An ungated bypass does saturate the policy (99% of eval steps with $|a|\ge 0.9$) and collapses return to $-195.5$. A gated bypass that fires only on the flat shoulder $|a|\in[0.9,0.999]$ also fails, and does so without leaving a saturated policy. Warm-started MetaDrive fine-tuning shows the same pattern: the bypass does not improve return, and where collision rate falls it is typically traded for out-of-road departures. Auto-tuned entropy coefficient rises against the bypass, which is a push toward the tails. The Jacobian effect is real. Treating it as a bug to be undone is not free, and on the tasks studied here it is not helpful. Saturating a bound is not the same as solving a problem whose optimum lives on that bound.

Execution feedback can guide coding agents toward correct repository repairs, but only when the tests capture the behavior requested by the issue. Agent-generated tests can encode incomplete or incorrect behavioral targets; when the same trajectory writes both the patch and the test, their errors can agree and create false confidence. We introduce ExecCritic, combining a test--verify--revise scaffold with a role-specific reinforcement learning recipe for training agents within it. The scaffold separates test construction from source-code repair: a Test agent independently generates repository-native tests, a fail-closed harness qualifies and freezes them, and a Repair agent revises source code from their execution feedback without changing the tests. Both roles use Qwen-3.5-35B-A3B as the backbone and are trained separately. In Learn to Test, the Test agent learns to produce behaviorally valid tests that distinguish correct from incorrect patches. In Test to Improve, the Repair agent learns both direct task resolution and feedback-guided revision. On SWE-bench Verified, test quality determines whether feedback helps: holding the base Repair agent fixed, tests from the base Test agent reduce resolved rate from a no-test baseline of 61.2% to 57.3%, whereas tests from GPT-5.6-sol raise it to 65.3%. Role-specific post-training raises the Qwen Test agent's Base-to-Gold success from 22.2% to 62.2%; composing the two post-trained Qwen agents reaches 72.6%, an 11.4-point gain over the original no-test baseline without stronger-model or Oracle feedback at evaluation time. Code is publicly available at https://github.com/MSR-Orchard/execcritic.

Model-based reinforcement learning (MBRL) is a family of RL methods that learn a model of the environment and use it for action selection, making it well suited to robotics due to its sample efficiency. Combining learned models with online planning can further improve action selection, as the planner can exploit the model to find better actions than the learned policy alone. Recent methods combining learned policies with online planning typically learn the value of the policy rather than the stronger planner-guided behavior. We present CAST (Critic with Alternating State-value Target), which uses planner-guided behavior to improve value learning while regularizing the value estimate with the current policy. CAST replaces the action-value critic with a state-value critic, trained using a target that combines a real planner-guided transition and an imagined transition under the current policy. The resulting value function corresponds to an alternating process between planner-guided behavior and the current policy, allowing it to benefit from the stronger planner behavior while being regularised by the policy being learned. We evaluate CAST on the DeepMind Control and HumanoidBench Suites against several state-of-the-art methods, and demonstrate successful transfer to a physical Unitree Go2 quadruped performing a dynamic handstand.

Formulaic alpha discovery is a pool-dependent symbolic search problem in which informative feedback is observed primarily when a complete expression is evaluated. This delayed feedback creates two coupled difficulties: the retained alpha pool does not preserve the full history of realized evaluation feedback, and the value of an intermediate construction action is uncertain because its consequence depends on the formula eventually completed. We introduce AlphaRJM, which addresses these difficulties through Reward-Jump Memory, an event-driven latent state that remains fixed during token construction and updates only at terminal evaluation events using the realized pool reward and evaluation outcome, and an action-conditioned SDE return critic that represents future discounted discovery returns with stochastic particles. The particles guide action selection through their mean and uncertainty and are learned using a distributional Bellman objective combining energy-distance matching, mean calibration, and jump regularization. Empirically, AlphaRJM delivers strong and stable gains across multiple equity universes, forecasting horizons, and random seeds, while ablations confirm the complementary roles of persistent evaluation history, stochastic return modeling, and distributional supervision.

Long-form narrative-to-film generation requires shot-level controllability and cross-clip consistency in both visual identity and character behavior-requirements that remain difficult to satisfy with current prompt-based workflows. A core reason existing workflows remain brittle is the lack of a structured intermediate layer between scripts and video models, especially when screenplays are underspecified at key cinematic decision points. We introduce a structured orchestration layer for film-oriented script-to-video generation, implemented as a multi-agent framework that operates between scripts and off-the-shelf video generators. The layer is centered on FilmDSL, a film-oriented domain-specific language that makes cinematic constraints explicit, including shot and camera directives, asset and continuity requirements, and persona cues, so that agents coordinate through a shared structured specification for planning, generation, critique, and repair. Specifically, a generation agent constructs asset packs and storyboard keyframes that anchor composition before clip-by-clip synthesis, while a critic agent produces structured QA signals and triggers targeted refinement without retraining the base model. Experiments on TV-style segments show improved controllability and consistency over text-only and reference-only baselines.

We introduce the first Probably Approximately Correct (PAC) learning framework for general-sum concurrent stochastic games (CSGs) with transition uncertainty, while addressing the challenge of Nash equilibrium (NE) existence. Our algorithm maintains data-driven $L^1$ confidence sets over transition kernels and solves a robust CSG to compute a social-welfare optimal $\varepsilon$-NE, using a robust MDP-based exploration mechanism to drive joint state-action coverage. Crucially, we introduce a Nash margin characterisation that enables principled reasoning about equilibrium existence: the framework either returns an $\varepsilon$-approximate NE whose social-welfare value is $\varepsilon$-close to optimal, or provides a sound certificate that no exact NE exists. Under a minimum reachability condition $p_{\mathrm{reach}}>0$ over relevant state-action pairs, the algorithm terminates after a polynomial number of trajectory samples, with sample complexity $\widetilde{O}\left( {R_{\max}^2 H^4 |S|^2 |A| / (p_{\mathrm{reach}} \varepsilon^2)} \right)$. Empirical results on benchmark CSGs demonstrate near-optimal performance, correct handling of equilibrium (non-)existence, and sample complexity consistent with theory.

Real-world robotic assembly at sub-millimeter tolerances demands spatial precision, compliant interaction, and robustness to contact failures. We present Facet-0, a robotic foundation model that predicts and values the contact consequences of its actions. Facet-0 unifies multimodal representation learning and reinforcement learning (RL) post-training around a joint action-wrench proposal: a causal wrench history is aligned with vision-language semantics and kinematic state, and flow matching generates each action chunk together with the future wrist-wrench profile it is expected to induce. Deployment rollouts train a distributional Action-Wrench Critic to distinguish motions with similar task progress but different contact outcomes, while phase-aware rewards and contact-selective credit concentrate policy improvement on decisive interactions. To accommodate part-specific dynamics, a lightweight bounded actor reuses the frozen representation for on-robot adaptation; RL remains defined over executable Cartesian actions, while an auxiliary wrench head preserves predictive, non-commanded action-contact coupling. Trained on ManuFacet-1K, a 1,000-hour force-synchronized corpus spanning three embodiments and multiple manufacturing cells, the bounded task-adapted system reaches 82% mean success on five sub-millimeter computer-assembly tasks, compared with 15% for the strongest baseline, with 0.5 mm placement accuracy and 50 ms command latency.

Vision-Language Models (VLMs) still struggle on tasks requiring complex visual understanding. We argue that the core issue is not high-level reasoning, but instead failing to locate critical details in the image. Due to this shortcoming, VLMs generate often plausible but incorrect reasoning based on flawed perceptual grounding. To address this, we propose Locator-Critic (LOCI), a training-free framework that decouples visual search from evidence verification. LOCI employs a Locator agent to propose candidate visual evidence and a separate Critic agent to evaluate its relevance and sufficiency. These agents engage in an iterative refinement loop, progressively improving the evidence until it is adequate to answer the given question. This decoupled, self-correcting process yields substantial performance gains, achieving state-of-the-art results on multiple complex visual benchmarks. LOCI improves accuracy for both open-weight models like Qwen3-VL (+12.1 on V*, +5.8 on HR-Bench and +11.2 on VisualProbe-Hard) and proprietary models like Gemini 2.5 Pro (+8.9 on V*, +4.3 on HR-Bench, +4.8 on VisualProbe-Hard).

Goal-conditioned reinforcement learning struggles with long horizons when rewards are sparse. While a planner can provide subgoals to guide a low-level policy, its use at test time may introduce practical subgoal management difficulties. An alternative paradigm utilizes a high-level planner to assist learning, while the policy remains conditioned only on the final goal, enabling planner-free deployment. Among these methods, Reinforcement Learning with Imagined Subgoals (RIS) introduces a regularization term that encourages the policy to take the same actions for the final goal as it does for an intermediate goal. This regularization, however, may lead to goal-chaining issues when intermediate goals are low-dimensional. Potential-based reward shaping (PBRS) translates plans into an additional reward while ensuring that the optimal policy remains unchanged. Yet, it can generate deceptive rewards in terminal states. We study these failure cases and first propose an alternative reward shaping method (RS) that removes these deceptive rewards at the expense of theoretical guarantees of PBRS. Similar to this RS variant, we then propose another method named Locally-Guided Actor Critic (LG-AC) that rewards the agent for reaching intermediate goals. Unlike RS, where intermediate rewards are implicit in the shaping signal, we explicitly condition a value estimator on the full sequence of intermediate goals but represent the value function as a sum of subgoal-conditioned value functions, enabling dense hindsight relabeling. We evaluate all these methods in tasks with challenging goal-chaining requirements and empirically highlight specific cases in which either action regularization or reward shaping yield low performance, while LG-AC achieves the best overall performance across tasks.

Natural-language critiques provide supervision beyond scalar rewards for non-verifiable generation, which lacks deterministic verifiers. In critique-guided refinement, a critic gives feedback on an initial response and an actor revises it. However, final revision quality does not reveal whether the critique was actually useful: a capable actor may improve without following the feedback, while valid feedback may fail if the actor cannot execute it. We frame critique as actor-conditioned revision guidance, where usefulness depends on whether the feedback helps the target actor address the intended weakness. We introduce TAIScore (Targeted Actionable Improvement Score), a reward that evaluates the instruction, initial response, critique, and revision together, assessing whether the critique targets a real weakness, whether the actor follows it, and whether the intended aspect improves. We use this reward to train an actor-tailored critic with GRPO, and use critique-guided refinements to construct DPO preference pairs for the actor, forming a co-evolving critic-actor loop where the critic adapts to the actor's changing capability. Experiments show that an 8B critic trained with TAIScore outperforms both a zero-shot 120B critic and critics trained with outcome-only or critique-only reward signals. Co-evolving the critic and actor further improves performance, suggesting that effective critique supervision should adapt as the actor changes.

Large language model (LLM) agents are increasingly deployed in long-horizon, interactive, and stateful environments. In these settings, a single wrong action, such as refunding the wrong purchase, can cause irreversible task failure and must be intercepted before execution. Such failures may not appear in every single run, but can emerge across repeated trials, making reliability across steps and trials critical. However, ensuring agentic reliability is challenging: even frontier LLMs struggle to explain why an action may be wrong, especially in long, intertwined trajectories governed by domain-specific policies. Much recent work relies on prompt-based critique agents, while optimization-based methods lack a systematic way to produce rich verification rationales for training. We address this gap with CAST, a critique-aware training framework that converts sparse task outcomes into action-level supervision for critique learning and policy optimization. CAST analyzes agent trajectories to synthesize structured rationales explaining action validity under partial observability. The resulting critique model is used to construct critique-aware training data for optimizing the policy model. Fine-tuning Qwen3-family models on dynamic tool-calling benchmarks, CAST improves reliability across domains, outperforming GPT-OSS-120B by over 10% pass^4 on Retail tasks and yielding an additional 9% improvement on Telehealth in an out-of-domain setting. These results demonstrate that critique-aware training improves the robustness of LLM agents in realistic dynamic environments.

Music recommendation relies primarily on two signals: user-item interactions, which fail in the cold-start regime, and intrinsic musical content, available for any recording. We argue that a third, largely untapped signal is both richer and more principled: critical adjacency, the pairwise relation established when an expert critic explicitly links two artists in long-form prose. It encodes deliberate judgments about which artists belong together. Prior work established its internal validity, showing it recovers coherent, interpretable communities and can match collaborative filtering in user-satisfaction simulations, with no user data. What has been missing is external validation: whether this critic-sourced relation is grounded in the music itself versus sociological context. We test it against acoustic content, reframing the question as one of construct validity. Representing artists as empirical distributions over 80 low-level Essentia acoustic descriptors and modeling pairwise proximity via marginal optimal-transport (Wasserstein) distances, we evaluate how far critical adjacency is sonically recoverable under a cold-start, artist-disjoint split. Our ensemble recovers these edges at out-of-sample AUC of 0.767 (95% CI 0.761-0.775). Recoverability rises monotonically with critical consensus, reaching 0.865 on multi-source attested edges. Stratified evaluations align with sociological models of genre: tightly bounded, scene-based genres show higher recoverability than broad industry umbrella terms. Critical discourse is thus a rich source of information for recommendation, decomposing into a reproducible "sonic core" and a "sociological remainder" driven by narrative positioning, subcultural context, and canonical placement. The work offers both a scalable cold-start discovery mechanism and a sociologically grounded approach to MIR and MRS research.

UAV on-board vision systems are widely used for different activities, including monitoring in no-fly zones. In this case, the vision-equipped UAV streams a video to a ground server where an operator assists its activities. The latency of video transmission has a profound impact on the effectiveness of the operator assistance. However, most techniques available for video transmission still incur significant latency costs. In this paper, we propose a graph convolutional neural network-assisted (GCN-Assisted A2C) deep reinforcement learning (DRL) system model to find the optimal pixel-correlated area of a suspicious object. We combine the Lagrangian dual form with gradient descent to prevent lack of convergence and over- and under-penalization constraint violation during latency optimization. The proposed system model sends a sub-group pixel-correlated area of the frame from the UAV to the server rather than the transmission of the whole video frame. The proposed framework utilizes the GCN model to explore hidden representations of feature-correlated groups of pixels. Moreover, the GCN supervises the A2C model, which selects a subgroup to enhance transmission latency, thus supervising the training of UAV actions in A2C. Experimental results show that GCN-assisted A2C reduces video frame transmission latency together with false detection rate in UAV vision systems over other DRL and state-of-the-art models.

Long-horizon surgical robot manipulation is challenging because task rewards are sparse, while meaningful interaction changes occur at irregular intervals. Existing world-model agents typically imagine at primitive-step resolution, leaving variable-duration task progress implicit. Manually specified stages can provide intermediate structure, but their task specific boundaries are difficult to align with state-dependent interaction transitions. We propose S2-HWM, a Sparse Event-Structured Hierarchical World Model that learns sparse event evidence from primitive latent trajectories to coordinate an event-level manager and a primitive-step worker. The event evidence schedules manager goal updates, and each selected latent goal conditions the worker's primitive actions until the next update. The learned event evidence also forms variable-duration segments for an Event Transition Model (ETM), which predicts the next?boundary stochastic state, segment duration, and accumulated segment reward. Chaining these event-level predictions provides a variable-duration continuation beyond the primitive imagination horizon for manager learning, while the worker retains primitive-step actor-critic learning. On a SurRoL-based PegTransfer task, S2-HWM achieves a success rate of 98.7%, outperforming the flat GAS DreamerV3 baseline by 22.7 percentage points.

This paper proposes a contextual quality-diversity evolutionary reinforcement-learning controller, CQD-ERL, for the supervisory control of a tropical, water-cooled chiller plant and its associated air side. Rather than converging to a single scalarised policy, the controller maintains a product archive of specialised policies indexed jointly by a data- driven operating context, a cluster of daily weather and load regime, and a context-invariant behaviour descriptor, filled by a gradient-free evolutionary operator and a soft-actor-critic policy-gradient operator that share one replay buffer. Every action is filtered through a deterministic safety shield before execution. The controller is trained on a two-tier reduced-order environment representing the latent load, cooling-tower approach and humidity constraints of a Singapore commercial building, and is evaluated over a full annual backtest against an ASHRAE Guideline 36 baseline.

Offline-to-online (O2O) reinforcement learning aims to leverage policies pretrained on static datasets while improving them through online interaction. However, directly reusing an offline-trained critic can hinder online fine-tuning: as the policy and data distribution change rapidly, value estimates inherited from offline training may become misaligned with the online environment, leading to inaccurate policy improvement and inefficient exploration. To address this problem, we introduce Critic-Free Pretraining: an efficient paradigm that completely abandons the approach of offline critic training, allowing a freshly initialized critic to adapt without inheriting biased estimates. CFP is compatible with various mainstream O2O algorithms and consistently matches or improves upon conventional O2O algorithms across a diverse set of tasks, with particularly pronounced gains on several challenging tasks.

Real-world online reinforcement learning (RL) provides a promising approach for training robotic manipulation policies directly in the physical world, avoiding the sim-to-real gap and enabling continuous policy refinement through human-in-the-loop interaction. Recent methods have demonstrated sample-efficient learning through human intervention but remain limited to small randomization ranges and encounter challenges with the non-stationarity induced by concurrently training multiple agents. To address these limitations, we introduce a unified framework that combines centralized training with decentralized execution (CTDE) and a Hybrid Reward Architecture (HRA). This enables multiple actors to share a centralized multi-head critic. The critic is decomposed into task and grasp heads, corresponding to the sparse task reward and a potential-based grasping reward, respectively. We accordingly reformulate the critic and actor objectives to exploit the decomposed Q-values while explicitly accounting for the categorical action distribution of the discrete gripper policy. Experimental results demonstrate that the proposed framework substantially improves both sample efficiency and policy performance. We validate our approach on two robotic arms and a simulated humanoid robot across tennis ball and banana pick-and-place, pot reset, and simulated block relocation tasks under dimension-wise domain randomization, approximately 5-25x larger than those considered in prior work. Compared with a state-of-the-art baseline, our method improves the success rate from 60% to 80% on tennis ball pick-and-place, from 60% to 90% on banana pick-and-place, and from 25% to 95% on simulated block relocation, while also successfully accomplishing a task where the baseline consistently fails. Videos and more details are available at our project website: https://hil-harc.github.io/.

On--off cycling is the main cause of compressor wear in residential heat pumps, yet reinforcement learning (RL) controllers for buildings typically optimise only energy cost and thermal comfort, ignoring how much the learned policy cycles. We add a levelised compressor-wear term to the control reward and study how the resulting behaviour depends on the RL algorithm. Training Soft Actor---Critic (SAC) and Proximal Policy Optimisation (PPO) on an identical Markov decision process for the BOPTEST bestest hydronic heat pump case, we find that SAC learns a continuous modulation policy that keeps the compressor permanently engaged---the operating principle of an inverter-driven heat pump---achieving zero start-ups per day, whereas PPO collapses to bang-bang control that cycles more than the baseline. On the BOPTEST emulator the SAC policy cuts thermal discomfort by up to 90.7% for an 11.5% cost increase, while eliminating all baseline cycling.

Improving sample efficiency remains a core challenge in reinforcement learning (RL), especially in real-world settings like robotics, where data collection is costly. This challenge is pronounced in visual RL, where high-dimensional inputs often obscure learning signals. While prior work in visual RL has focused on algorithmic solutions, such as better dynamics models or exploration strategies, recent advances in state-based RL show that architectural design alone can lead to significant gains in sample efficiency. This raises an important question: Can these architectural principles transfer to visual RL? In response, we introduce V-Simba, a simple yet effective visual RL architecture inspired by the Simba architecture from state-based RL. Built on top of Soft Actor-Critic (SAC) with data augmentation, V-Simba modifies the architecture by adding normalization layers to stabilize training and using pointwise convolutions to reduce computation. Despite its simplicity, V-Simba matches or outperforms the state-of-the-art methods across the DMC, Adroit, and Meta-World benchmarks, while being more computationally efficient than DrQ-v2. We make our code publicly available at https://github.com/DAVIAN-Robotics/V-Simba.

Hardware failures require legged robots to rapidly reorganize coordination and gait timing to maintain stability and mobility. This is particularly challenging for larger quadrupeds, where increased mass and tighter actuation limits reduce the feasibility of aggressive, high-frequency compensation strategies often observed on smaller platforms. In this work, we propose a deep reinforcement learning approach for fault-tolerant locomotion under actuator power loss. The method employs an asymmetric actor-critic architecture in which the critic has access to privileged information during training, while the actor learns to reconstruct a corresponding latent representation from proprioceptive observations. We introduce a latent-alignment loss that encourages consistency between actor and critic representations. Additionally, we augment the action space with a learnable gait frequency parameter, enabling adaptive gait timing in response to terrain variations and actuator degradation without predefined faulty-leg strategies. The approach is validated in high-fidelity simulation on uneven terrain and real-world experiments on flat ground using a 68 kg quadruped robot.

When a reinforcement learning agent cannot observe the full state, we usually blame its policies: it cannot see enough to represent a good one. We show that in a solvable case the bigger problem lies elsewhere. Even when a good policy is available and the agent's value function is expressive enough to describe it exactly, learning still ends up somewhere far worse. We study a partially observed linear-quadratic problem in which a standard actor-critic learner can be solved in closed form. At our default setting the best policy the agent can represent is already close to optimal, costing 10.4% more than the ideal controller that observes everything. Learning does not find it. The algorithm instead comes to rest at a policy that is 35% worse than the best one available to it, and we can say exactly where and why. The cause is a bias in what the critic learns rather than a limit on what the actor can express. Because the agent cannot attribute what it sees to the part of the state it cannot observe, the critic misreads that unexplained variation as sharp curvature in its own value estimates, and the actor follows that error away from the optimum. We derive closed-form expressions for the resulting policy, for its cost, and for the one design choice that removes the problem, which is how far the learner looks ahead before trusting its own value estimates. Deep reinforcement learning experiments follow these predictions closely. Notably, giving the agent memory of past observations does not help, while changing how far it looks ahead does.

Reinforcement learning with Verifiable Reward (RLVR) has emerged as a powerful paradigm for training coding agents, where the execution feedback from compilation and tests provides objective verification. However, unlike agent tasks, coding agents face a unique and finer-grained credit assignment challenge: at each step, coding actions simultaneously pack varying changes into different regions of a code version, which makes the contribution of independent change indistinguishable. Existing RLVR methods mostly leverage the outcome reward or step-level reward, which fails to dive into a code diff and makes unique properties of coding actions invisible to training. In this paper, we propose Diff-in-Diff Policy Optimization (DiDPO), a critic-free RL method that constructs fine-grained credit units directly from the structure of code diffs. DiDPO organizes multi-turn coding interactions into multiple thought--action steps and discovers code diffs across sampled trajectories. It then selects anchors by aggregating highly similar sub-diffs split from each whole diff by our ``groupability score'', which provides the splitting schema that optimally balances the semantic scope of anchors and the group mass they may form. Finally these anchors form advantage groups and project the diff-level advantage back to individual response tokens. Experiments on long-horizon coding and reasoning benchmarks show that DiDPO significantly outperforms strong agentic RL baselines. On Qwen2.5-7B-Coder, DiDPO exceeds comparable methods by over 10% and narrows the gap with far larger models, offering a principled framework for fine-grained credit assignment in coding agent training. We also open-source verl-code, an agentic rl codebase that supports various RL methods and coding benchmarks.

Critic-free group-based reinforcement learning has become a scalable approach for post-training large language models. However, most existing methods allocate the same number of rollouts to every task and trajectory state, even though some rollouts provide much more useful learning signals than others. Recent work has started to treat rollout generation as an adaptive decision, but two important limitations remain. First, intervention strategies are often based on fixed heuristics and therefore cannot adjust as the policy changes during training. Second, these methods usually decide only how many rollouts to generate, without explicitly controlling where and how to intervene. To address these limitations, we propose Recoverability-Aware Intervention Learning (RAIL), a training-time framework that learns how to generate rollouts based on the improvement produced by each intervention. RAIL models intervention selection as an online contextual-bandit problem and trains a recoverability controller using intervention traces collected through a shadow-to-live procedure. This allows the controller to keep learning while the underlying policy evolves. We evaluate RAIL in terms of effectiveness, adaptivity, expressiveness, and efficiency. Across multiple settings, RAIL consistently improves performance under limited rollout budgets. These results show that recoverability-aware intervention provides a principled way to generate more informative and less redundant rollouts, leading to stronger learning signals during post-training.

Safe actor-critic control often treats barrier filtering, uncertainty estimation, and experience replay as separate modules, even though each changes the data used for learning and control. We develop an integrated architecture in which the uncertainty estimate updates the obstacle geometry used by a control barrier function, filter interventions and estimation residuals determine replay priority, and the critic learns from the executed rather than nominal action. We instantiate the architecture on a two-dimensional robot-navigation task with corrupted obstacle measurements and compare six component-matched configurations under common training budgets, random seeds, sensor streams, exploration, and disturbances. Evaluation includes a moderate post-training test, an eleven-level perception-noise sweep, and an exploratory extreme-stress test at multiplier $6.0$. In the extreme test, the integrated configuration recorded no contacts and reached the goal in all five evaluation seeds. Its mean cost was $7.63\pm0.44$ and its obstacle-belief root-mean-square error was $3.52\pm0.55$ cm. The uncertainty-estimation ablation also recorded no contacts but reached the goal in four of five seeds, with mean cost $8.96\pm2.08$ and belief error $11.08\pm1.23$ cm. A finite-training bound clarifies replay exposure, and a robust barrier condition states the required estimation-error and feasibility assumptions. The results support coupling estimation, safety filtering, and replay on this benchmark; broader safety and convergence claims require further study.

Vision-Language-Action (VLA) policies trained by behavior cloning fail silently: from the action stream alone, a collapsing rollout looks much like one making clean progress, because imitation supplies no notion of progress. Reinforcement learning would supply one, but it is impractical here, where real-robot experience is costly and deformable food resists simulation. The cheap alternative, a terminal success / failure bit, is learnable in principle yet far too sparse to say when a rollout went wrong. We argue that the per-frame label, not the architecture, is the hard part: to be useful it must be dense, continuous, and correctly shaped. We present ValueFormer, a compact policy-agnostic causal transformer over a frozen DINOv3 backbone that emits two per-frame signals in one forward pass: a smooth Monte Carlo value, V_mc, for advantage estimation and a sharp binary value for online mistake detection, targets that pull in opposite directions by design. Failed episodes are labeled with a stage-aware, success-then-decay return that preserves the success curve before the failure stage, and detection is supervised from mistake intervals rather than a single failure time, so mistakes the policy recovers from also carry signal. On a real-robot bimanual sandwich-assembly task 1,427 episodes), a critic-derived per-frame training weight lifts task completion from 70% to 85% (within noise at n=20), and a batched bf16 encoder cuts the live serving cost 3~5 times so the critic runs at 2 Hz alongside the policy on a single GPU.

Reinforcement learning and control theory are two adjacent scientific fields that focus on optimizing the controller of unknown dynamical systems using feedback. While both fields have common roots in dynamic programming, they have evolved with distinct methodologies, goals, and cultures. Despite decades of mutual influence, a significant gap persists between the two communities. This tutorial introduces adaptive control, actor-critic reinforcement algorithms, and a new way to combine these two paradigms for data-driven decision making on a classical locomotion control problem. Our aim is to provide a foundation for understanding the core differences between the two approaches and insights to help experts in each field better understand and engage with the tools and approaches of the other.

Proximal Policy Optimization (PPO) for large language models typically trains its critic by mean-squared-error (MSE) regression on scalar value targets. Although scalar MSE is statistically valid for estimating the conditional expected return, sparse binary rewards in reinforcement learning with verifiable rewards (RLVR) make critic optimization and calibration especially consequential: small value errors directly distort the scalar advantages used by PPO. We study whether a classification-based training objective can improve this critic signal. HL-Gauss PPO replaces the scalar MSE head with a categorical predictor over a discretized value support, trained by cross-entropy against smoothed HL-Gauss targets. Its output is decoded to a scalar expectation for standard GAE and PPO; the actor update is therefore unchanged and is not distributional. Across mathematical reasoning, tool-augmented math, and Search-R1, and on both Qwen2.5 and Qwen3 backbones, HL-Gauss PPO consistently improves over strong PPO and DAPO baselines. Controls with one-hot, two-hot, and Bernoulli two-bin critics show that neither a larger output head nor binary classification alone explains the gains. On a common collection of reasoning prefixes, HL-Gauss improves Brier score and calibration error and yields more symmetric, lower-variance advantages. These results position categorical value learning as an effective optimization surrogate for PPO critics in RLVR.

Reinforcement learning (RL) post-training of Vision-Language-Action (VLA) models has shown strong promise for robotic manipulation. Among RL methods, critic-based approaches rely on a value estimator that predominantly operates on single-frame observations or single-frame VLM backbone latents, which is a fundamental mismatch with the partially observable nature of robot control. A naive approach to incorporate observation history into the critic incurs exponential complexity with high-dimensional visual space, and still fails because pure scalar-return regression provides insufficient supervision for learning cross-temporal dynamics. We identify the root cause as a state approximation problem: without an explicit world modeling objective, the critic's representation cannot capture the temporal structure needed for accurate value estimation. To address this, we propose the World Critic Model (WCM), built on a lightweight LeJEPA architecture; WCM jointly predicts future latent state and estimates values, such that the critic's representation is explicitly trained to capture temporal dynamics rather than merely regress scalar returns. WCM integrates seamlessly into both on-policy and off-policy training pipelines and is compatible with state-of-the-art VLA backbones including Pi0, Pi0.5, and OpenVLA-OFT. Extensive experiments on 149 tasks across four benchmarks demonstrate that WCM consistently achieves state-of-the-art performance in both in-distribution and out-of-distribution settings, with particularly strong generalization gains. We further validate WCM on seven real-world manipulation tasks using OpenVLA-OFT and Pi0.5 with off-policy RL, confirming stable deployment across diverse settings.

Constrained Markov Decision Processes (CMDPs) provide a natural framework for reinforcement learning in safety-critical applications, where agents maximize long-term reward while satisfying long-term constraints. Although primal-dual actor-critic methods with linear critics are well understood, extending order-optimal convergence guarantees to neural critics in average-reward CMDPs has remained open. The main challenge is a fundamental bias-cost trade-off in neural critic estimation: under Neural Tangent Kernel (NTK) analysis, reducing critic bias substantially increases critic optimization cost, preventing order-optimal convergence in the primal-dual framework. We resolve this bottleneck by introducing a hierarchical Multilevel Monte Carlo (MLMC) neural critic that performs debiasing simultaneously across trajectory sampling and critic optimization. The resulting estimator attains the bias of a long critic optimization run with only logarithmic expected sample cost. Building on this estimator, we develop a primal-dual Natural Actor-Critic algorithm that achieves both an optimality gap and a constraint violation of order $\tilde{O}(T^{-1/2})$. This establishes the first order-optimal convergence guarantees for infinite-horizon average-reward CMDPs with general policy parameterization and neural critics, while eliminating the need to know the underlying mixing time. Our results are novel even in the unconstrained setting.

Reinforcement learning has emerged as an effective paradigm for enhancing the mathematical reasoning capabilities of large language models. Among existing policy optimization methods, Proximal Policy Optimization (PPO) remains particularly appealing because its learned critic can, in principle, provide token-level credit assignment. However, in mathematical reasoning tasks characterized by long reasoning horizons and sparse outcome rewards, reliable token-level credit assignment remains challenging. The standard critic often fails to accurately evaluate intermediate reasoning states, resulting in noisy advantage estimates and suboptimal policy updates. In this paper, we propose ReDiPPO, a Reference-guided and Discrepancy-aware PPO framework for mathematical reasoning. ReDiPPO introduces a reference-guided critic that uses reference answers as training-time privileged signals to provide more accurate value estimation. Meanwhile, it retains a standard critic and quantifies the token-level reference-standard discrepancy between the standard value estimate and the reference-guided value estimate. This discrepancy serves as an indicator of difficult reasoning states and is used to reweight the corresponding token-level advantages during PPO optimization. Extensive experiments on diverse mathematical reasoning benchmarks demonstrate that ReDiPPO improves value-estimation accuracy and consistently outperforms strong policy optimization baselines, including PPO, DAPO, and GSPO, in final reasoning performance. Our code is available on https://github.com/cii030/ReDiPPO.

Deep off-policy reinforcement learning algorithms for continuous control typically rely on neural value function approximation to guide policy improvement. However, temporal-difference (TD) learning introduces noisy targets, resulting in non-stationary optimization, while greedy policy updates amplify early-stage estimation errors. The recursive propagation of such errors leads to persistent overestimation bias and degraded training stability in actor-critic methods. Existing approaches attempt to alleviate this issue via prioritized sampling or modified value learning objectives, but often overemphasize high-uncertainty transitions caused by limited data coverage or bootstrapping errors, thereby further amplifying bias.In this paper, we propose Collaborative Weighting Actor-Critic (CWAC), a unified framework that explicitly accounts for predictive uncertainty in value estimation. CWAC employs distributional critic to model return uncertainty and introduces a collaborative weighting mechanism that jointly reweights TD-errors and uncertainty, enabling robust learning from reliable samples while suppressing noisy updates. In addition, we incorporate a stochastic pessimistic value estimation scheme via sampling from the return distribution, which effectively mitigates error propagation during policy improvement. CWAC can be seamlessly integrated into existing off-policy algorithm frameworks such as SAC, TD3, and DDPG with minimal overhead. Empirical results demonstrate that our proposed method significantly enhances performance across a diverse range of simulated tasks. Our code is publicly available at https://anonymous.4open.science/r/CWAC-348E.

This paper presents a cooperative indoor UAV guidance framework that combines a shared voxel-map world model with a multi-agent Soft Actor-Critic (MASAC) controller. Multiple drones fuse 360 LiDAR observations into a common world-frame occupancy map, which is converted into a compact bird's-eye-view (BEV) representation and provided to each agent as an ego-aligned local crop. This integrate-in-world, act-in- ego design enables consistent multi-UAV spatial fusion whilst retaining decentralised continuous control. The policy combines BEV map features, near-field obstacle observations, and compact goal and peer-state information within a centralised-training, decentralised-execution framework. In simulation, the learned controller achieves a 90.3% success rate in corridor navigation, outperforming Astar planning, an artificial potential field controller, and a prior guidance method. To address residual sim-to-real mismatch, the simulation-trained policy is further adapted using offline imitation fine-tuning from real-world data. Real-world experiments in GNSS-denied indoor environments demonstrate stable two-UAV cooperative operation across increasingly chal- lenging obstacle layouts. The results show that shared voxel-map representations provide an effective and scalable spatial substrate for learned cooperative indoor UAV guidance.

Exploration in sparse-reward long-horizon tasks poses significant challenges for reinforcement learning. To address these challenges, we propose a two-level Hierarchical Reinforcement Learning (HRL) framework. The first level handles high-level strategic planning, while the low-level uses the continuous-control Soft Actor-Critic (SAC) algorithm, and they utilize entropy-regularized policy optimization. The proposed framework was trained and evaluated using the Search-and-Rescue-2 (SAR-2) dataset. HRL-SAC effectively addresses sparse-reward long-horizon search problems characterized by delayed rewards and continuous control, and its outperforming the flat SAC baseline reinforcement learning in terms of success rates, coverage efficiency, and convergence. These findings indicate that hierarchical entropy-regularized policies are a promising solution to tackle long-horizon sparse-reward reinforcement learning tasks.

Multimodal large language models exhibit capabilities on reasoning tasks, yet often produce flawed intermediate steps while yielding correct final answers. This behavior undermines interpretability and reliability, suggesting reliance on spurious shortcuts rather than faithful reasoning. Although efforts have explored step-level supervision, distinguishing decisive steps from redundant ones remains challenging. We propose $O^2$-CritiCuRL, a novel curriculum reinforcement learning framework that introduces critical-step awareness through an iterative offline-online paradigm. In the offline stage, $O^2$-CritiCuRL conducts multi-rollout analysis over step-annotated trajectories to estimate step-level importance, allowing the framework to distill critical reasoning steps and filter out redundant ones. In the online stage, we employ a progressive step-level reinforcement learning strategy, where truncated chains guide the model to infer missing steps and refine its reasoning, thereby sharpening its focus on critical steps and overcoming the limitations of static supervision. Extensive experiments on multimodal reasoning benchmarks show that our method achieves state-of-the-art performance while delivering superior training and inference efficiency. Code is available at https://github.com/kk0013/CritiCuRL.

Large Vision-Language Models (VLMs) now act as agents in interactive environments, where success requires coherent reasoning and decision-making across turns. Although end-to-end training in agentic environments can improve such multi-turn decision-making abilities, current methods mainly rely on either token-wise optimization over concatenated token trajectories or turn-wise optimization with uniform within-turn credit. In this work, we establish theoretical formulations for the two levels of optimization and derive a hybrid advantage that serves both objectives. Furthermore, with an appropriate choice of discount factor and learning target, we prove that a unified critic model can estimate values for both turn-wise and token-wise. As such, we propose HyGAE, an actor-critic framework that jointly optimizes token- and turn-level objectives with the hybrid advantage and unified critic. We conduct extensive evaluations of HyGAE across five multi-turn decision-making environments, where it achieves an average success rate of 91% and a significant improvement of 10% over other methods. Furthermore, we provide an in-depth analysis showing that the exact analytic form of the hybrid advantage and return is crucial for optimization. Project Page: https://wx-zhang.github.io/hygae-web/.

Large language models increasingly write both code and the tests meant to check it; coverage records what ran, not what was verified. We study an adversarial test-hardening loop under a mechanical oracle: a Tester model writes tests, mutation testing names surviving injected defects, and a Critic model writes tests to kill exactly those, with every verdict decided mechanically, so no model judges another's output. In Experiment 1, on five Python subjects (one same-lineage-loop cell could not be scored), the loop killed 105 mutants that one-shot generation missed and lost none, and the cross-lineage-Critic question returned a pre-declared null. The central finding was an autopsy: an earlier analysis reported a cross-lineage effect at p = 9.5e-66 that was an instrument artifact, an output cap silently truncating the verbose model, caught only by adversarial review of the completed analysis. Review then found a further confound, each arm resampling its own initial suite; Experiment 2 removes it. Under a pre-registered frozen-shared-round-0 design (five replicates on each of four subjects, seeds committed in advance), same-lineage Critic rounds killed 78% of the survivors the frozen initial suite left standing (mean incremental kill rate 0.783, 95% cluster-bootstrap interval [0.592, 0.935]), a within-replicate causal estimate; the cross-provider configuration showed a positive pilot difference (rate gap 0.178, 95% interval [0.039, 0.347]; magnitude dominated by a single replicate) at 5.5x lower arm cost. This compares two named model-provider-harness configurations, not an isolated lineage effect: part of the gap is one configuration's receipted operational failures, including truncation recurrences, now detected and scored rather than laundered. Cross-model comparisons can inherit the asymmetries of the harness that runs them. We release both protocols, all receipts, and the analysis code.

This paper addresses defensive blocking in autonomous racing, where a vehicle must prevent a faster opponent from overtaking while operating near its dynamic limits. Different from lap-time minimization, we formulate defense as a spatial occupancy regulation problem via a hierarchical reinforcement-learning guided model predictive control framework. A Soft Actor-Critic strategic layer operates in the Frenet domain to generate geometry-aware defensive references, which are embedded into the nonlinear model predictive control formulation as spatial regularization under friction constraints. Evaluated on the Thunderhill West circuit in simulation, the framework increases average overtake time from 8.8 s to 14.6 s while significantly reducing opponent progress. Meanwhile, it allows the vehicle to utilize 83.4% of available tire force. The framework achieves a 33.3 ms mean solve time (13.9 ms std), supporting real-time high-speed adversarial interaction.

Small language models and coding agents increasingly generate web front-end code, yet their outputs are typically evaluated primarily for functional correctness. A generated interface may compile, render, and pass unit tests while still violating established interface quality principles, including accessibility barriers, deceptive design patterns, poor visual hierarchy, and excessive decision complexity. Existing auditing approaches face a trade-off between cost, coverage, and scalability: expert human review provides rich judgment but is slow and expensive; frontier vision-language models offer broader reasoning capabilities but remain costly to deploy at scale; and rule-based tools such as axe-core and Lighthouse are inexpensive but primarily capture mechanically checkable accessibility issues. We investigate whether a lightweight vision-language model can serve as an effective critic for generated interfaces. We unify 19 interface-quality principles from three complementary sources of HCI knowledge: WCAG 2.2 accessibility standards, deceptive design taxonomies, and established theories of perception, cognition, and interaction. To train this critic, we construct a verified dataset of approximately 10,000 generated web pages by synthetically injecting known violations into clean, LLM-generated Tailwind pages. Continued reinforcement learning on a 4B vision-language model improves micro-F1 from 36% to 84%, with 13 of 19 principles exceeding 80% F1. The resulting critic can audit generated interfaces, filter low-quality interface training data, and provide a reward signal for design-aware code generation. We release our data-generation recipe and injection/verification prompts to support reproducible evaluation and future work on scalable interface-quality assessment.

Production LLM-based financial sentiment analysis faces a structural cost trap: most queries are trivially classifiable, yet expensive cloud reasoners process them all, and the bill scales linearly with user count. We present TriAgent, a multi-agent committee stratified by contextual granularity -- a word-level lexicon (VADER), a sentence-level domain transformer (FinBERT), and a cross-sentence reasoner (Qwen2.5, 0.5B-14B-4bit, with Mistral-7B and Phi-3.5-mini cross-family checks). A three-way Semantic Divergence Index (SDI) measures pairwise disagreement across granularities and routes each query accordingly. Our central finding is the critic plateau: when the LLM is re-tasked as a critic over the smaller agents' outputs, F1 plateaus at ~0.87 across 1.5B-7B Qwen (bootstrap 95% CIs overlap), while a same-size 3-persona vote drops to F1=0.66, which is driven by granularity-stratified diversity. Three corollaries follow from the same SDI signal: (i) a Shared Consensus Dictionary on multilingual sentence-BERT answers 95% of Chinese queries from an English cache at F1=0.99 -- cross-border canonicalization at zero marginal cost; (ii) SDI doubles as a post-hoc LLM-hallucination detector at AUC=0.90; (iii) the SDI single-stage strategy attains the best risk-adjusted return (Sharpe=3.50) on a 20-ticker back-test, dominating both always-FinBERT (1.36) and always-LLM (0.11). At 10M-user scale, TriAgent saves $9.3M/year vs. a GPT-4o-mini baseline. Code, lexicons, and the SCD are released.

Parameterized action reinforcement learning has shown strong performance in environments requiring both discrete action selection and continuous parameterization. Prior work established the effectiveness of single-agent actor-critic algorithms - Greedy Actor-Critic (GAC), Soft Actor-Critic (SAC), and Truncated Quantile Critics (TQC) - on benchmark parameterized action tasks, but their extension to multi-agent settings remains largely unexplored. This paper presents a comparative study of shared-experience multi-agent extensions of these algorithms: Multi-Agent Greedy Actor-Critic (MAGAC), Multi-Agent Soft Actor-Critic (MASAC), and Multi-Agent Truncated Quantile Critics (MATQC). Rather than following the centralized training, decentralized execution (CTDE) paradigm, the proposed framework uses multiple independent actor-critic agents that share a replay buffer while maintaining separate policy and value networks. We evaluate the algorithms on the Platform-v0 and Goal-v0 benchmarks against their single-agent counterparts, using three-, five-, and ten-agent configurations to assess scalability. Performance is measured by average evaluation return and training time across ten independent runs, with one-way ANOVA and Tukey HSD post-hoc tests used to assess statistical significance. Results show that the multi-agent framework consistently improves Greedy Actor-Critic performance, while MASAC and MATQC show comparatively modest gains over their single-agent versions. Increasing the number of agents beyond five yields limited additional performance while substantially raising computational cost, particularly for MAGAC. These results highlight a trade-off between learning performance and computational efficiency, offering insight into the scalability of shared-experience multi-agent actor-critic methods for parameterized action reinforcement learning.

Flow policies can represent multimodal action distributions for robot manipulation, yet a robot must execute one action at each control step. When several proposals are sampled, critic-based ranking makes data collection depend on value estimates over candidate actions that may be weakly represented in replay. We introduce HCPG-Flow, an analytic rollout-time selector that augments SAC-Flow with hierarchical, object-centric contact-progress guidance while preserving its actor and critic objectives. HCPG switches from end-effector approach to task progress after contact, scores each proposal by the first-order reduction of a task-relevant distance, standardizes scores within the candidate set, and executes a temperature-controlled action embedding. Across ten simulated tasks, HCPG improves mean success over SAC-Flow on both benchmarks, including a 9.5 percentage-point gain on Maniskill. Four physical tasks further show high success with a 17.4% reduction in successful completion steps.Project page: https://hitxraz.github.io/HCPG-Flow/

Reinforcement learning is increasingly being considered for controlling real-world systems, from fusion plasma and autonomous vehicles to drug discovery and drinking water treatment, where reliability is essential and tuning budgets are limited. Actor-critic algorithms share a set of design decisions, such as how the policy is updated, how it represents the distribution over actions, how its gradient is estimated, and how often it is updated relative to the value estimator. Using a control task derived from a real water treatment plant, we analyze over 33,000 experiments to determine how these components affect variability across runs and sensitivity to hyperparameters. Common defaults, such as Gaussian action distributions with pathwise gradient estimators, are among the least reliable configurations, whereas bounded distributions with adaptive update schedules remain robust across a wide range of settings. These findings offer empirical guidance to practitioners across scientific and engineering domains for understanding and making component-level decisions when adapting actor-critic methods to new real-world control settings.

LLMs can perform language-based quantitative prediction from unstructured inputs, but remain susceptible to hallucinations and overconfident errors, making it critical to know not only what a model predicts, but when its predictions can be trusted. We introduce CARE-PPO, a reinforcement learning framework that establishes a connection between loss prediction for uncertainty estimation and actor-critic PPO fine-tuning, enabling joint learning of accurate numerical estimates and reliable confidence signals in language-based quantitative prediction. CARE-PPO uses a Confidence-Aligned Reward for Estimation, defined as a function of prediction error, to provide dense error-aware feedback to the actor while inducing the critic to learn a value function aligned with prediction quality. During inference, we repurpose the critic as a confidence estimator. Across two real-world tasks in healthcare and finance and two Qwen-3 model scales (4B and 8B), CARE-PPO achieves strong quantitative prediction performance, while producing significantly better-aligned confidence estimates through the critic than logit-based and verbalized baselines. These gains persist under realistic out-of-distribution settings across domains, spanning linguistic and domain shifts. Finally, CARE-PPO reduces task-specific overfitting on general instruction-following prompts, consistent with the broader generalization advantages of RL fine-tuning over supervised approaches.

LLM agents act in external environments where each action changes the state that later decisions condition on, and where a single wrong step can waste interaction budget or trigger irreversible side effects long before the final failure is observed. Reliable deployment therefore requires \emph{step-level confidence estimation}: a calibrated probability that each proposed action is productive, available \emph{before} the action is executed. Existing LLM confidence estimators are designed to score a response from the given prompt, but agent confidence also depends on execution consequences: whether similar actions in similar situations actually advanced the task after the environment responded. We introduce the \method (\methodshort), a self-evolving critic framework in which an LLM critic accumulates evidence from its own past judgments and their observed consequences. After each trajectory, a hindsight LLM that sees the full execution feedback votes on whether each step was productive. The resulting pseudo-labels populate a memory bank from which related productive and unproductive experiences are retrieved into the critic's prompt whenever a similar step recurs. \methodshort requires no training and uses no ground truth step labels. Across three agent benchmarks and three critic backbones, \methodshort attains the best calibration (ECE and Brier) and ranking (AUC) in every dataset--critic combination, reducing ECE by up to $54\%$ relative to the strongest training-free baseline.

This paper introduces Actor as Its Own Critic, a unified reinforcement learning framework, Cycle Group Relative Policy Optimization (CycleGRPO), that jointly optimizes region understanding and localization for Multimodal Large Language Models (MLLMs). Unlike existing separate pipelines, we leverage the inherent duality between the two tasks to construct a self-evaluating reinforcement learning paradigm: "region $\to$ text $\to$ region''. Specifically, a single MLLM first acts as the actor to generate region captions, then immediately transitions to a critic to ground its generated text back in the spatial domain. Therefore, CycleGRPO requires only region inputs, e.g., masks or bounding boxes, entirely bypassing the need for textual ground truths. A quality-aware token-level cycle-consistency reward is employed to assess the semantic discriminability of text captions via their physical localization accuracy. Empirically, built upon SAMTok, our CycleGRPO framework successfully bootstraps both capabilities simultaneously. Without any task-specific fine-tuning, the framework yields consistent performance gains across a wide range of benchmarks, including region captioning, region VQA, grounded dialogue, and referring segmentation. Overall, CycleGRPO offers a straightforward and scalable way to advance pixel-level capabilities in MLLMs. Code and models are released at https://github.com/devinxzhang/CycleGRPO.

This paper develops a model-free reinforcement learning framework for continuous--time extended mean field control problems, where both the dynamics and reward may depend on the joint distribution of states and controls. We adopt deterministic feedback policies, under which the state--action distribution is induced directly as a push--forward of the state law. This avoids optimization over stochastic kernels and bypasses key limitations of existing approaches in extended mean field settings. We first establish a model--free sensitivity formula for parameterized McKean--Vlasov dynamics and use it to derive a deterministic policy gradient formula expressed through an advantage--rate function on the Wasserstein space. We then refine this formula by introducing local value and advantage--rate representations that depend on the state, action, and joint state--action distribution, yielding a policy gradient that includes both action derivatives and measure--derivative terms with respect to the control distribution. These characterizations lead to a martingale--based learning principle and motivate a continuous--time deep deterministic policy gradient algorithm combining particle approximations, measure--dependent neural networks, temporal--difference learning, and exploration in either action or parameter space. Numerical experiments on stochastic Cucker--Smale consensus control and optimal liquidation with trade crowding demonstrate the efficiency, stability, and robustness of the proposed method, including problems with explicit dependence on the control distribution.

Precision industrial contact manipulation requires reliable robot policies under pose perturbations and contact-force constraints. Vision-language-action models offer broad generalization but often introduce high inference latency and GPU-memory cost, while vision-action chunking policies are more suitable for real-time industrial control. However, these policies are usually trained by behavior cloning and suffer from distribution shift in contact-rich tasks. This paper proposes PAC-ACT, a reinforcement-learning post-training framework for pretrained Action Chunking Transformer policies. PAC-ACT reformulates policy optimization at the chunk level, constructs an ACT-transferred actor-critic architecture, and introduces a hybrid behavior-prior constraint to preserve the pretrained action distribution during online fine-tuning. Experiments on industrial precision-contact benchmarks show that PAC-ACT improves task success, contact stability, and force safety while retaining low latency and low GPU-memory usage. On the Contour task, PAC-ACT significantly reduces peak contact force and decreases the proportion of force readings above 60 N by 46 times. Sparse-reward ablations further show that the proposed behavior-prior constraint enables effective exploration under randomized initial poses.

Large language models (LLMs) have shown increasing promise in solving open problems in mathematics. However, their performance can be further improved through agentic workflows tailored to real-world mathematical practice. To this end, we introduce ProofCouncil, a mathematical agent that is designed to tackle open problems using an author-critic architecture. ProofCouncil served as a submission to the second batch of FirstProof, a challenge consisting of 10 real-world mathematical problems that agents must solve autonomously. Its submissions for 6 of the 10 problems were judged by the referees to be correct up to at most minor revisions, showing the best performance among participating teams. We also evaluate ProofCouncil on 30 open problems collected from mathematical researchers. Among the 21 solutions that received human feedback, 5 were judged completely correct, 2 more were judged promising pending final verification, and a further 8 contained useful partial progress. In this short paper, we describe the development of ProofCouncil and the agent-building library used to create it, which we release as open source to the community.

We study timestep allocation for score-based diffusion sampling, where a learned reverse-time dynamics is discretized on a finite grid. Uniform and hand-crafted schedules are standard choices, but they rely on fixed prescriptions and can therefore be suboptimal. To address this limitation, we propose Adaptive Reparameterized Time (ART), a continuous-time control formulation that learns a time change by treating the speed of the sampling clock as the control, so that a uniform grid on the learned clock induces adaptive timesteps in the original diffusion time. Based on a leading-order Euler error surrogate, ART provides a principled objective for allocating timesteps along the sampling trajectory. To solve this deterministic control problem, we introduce ART-RL, an auxiliary randomized formulation with Gaussian policies that turns schedule learning into a continuous-time reinforcement learning problem. We prove that the randomized ART-RL formulation is equivalent to ART at the optimizer level, in the sense that its optimal Gaussian policy recovers the optimal ART time-warping rate through its mean. We further establish policy evaluation and policy improvement characterizations and derive trajectory-based moment identities that yield implementable actor--critic updates for learning the schedule. Across experiments ranging from controlled low-dimensional settings to image generation, ART-RL can be plugged into existing diffusion samplers by changing only the timestep grid, consistently improving sample quality over strong baseline schedules at matched budgets while leaving the rest of the sampling pipeline unchanged. The learned schedules also exhibit broad generalization, transferring without retraining across sampling budgets, datasets, solvers, pipelines, and representation spaces.

Flow-matching vision-language-action policies generate robot action chunks through an iterative transport process, creating an opportunity for test-time guidance without retraining the base policy. We study this opportunity in Guided Action Flow, an inference-time framework that keeps a pretrained SmolVLA policy frozen and uses a learned action-chunk critic to guide its reverse-time flow sampler. The critic is trained from real success and failure rollouts, can condition on task-description features from the frozen SmolVLA language pathway, and is used only through action gradients during sampling. We evaluate the approach on LIBERO manipulation tasks. A single-task critic improves success from 68.0% to 82.0% on one seed window and from 82.0% to 86.0% on another. A multi-family task-description critic improves validation success from 46.0% to 56.0%, while the locked held-out test gain is positive but modest, from 65.0% to 67.5%. These results support the feasibility of Q-guided inference for frozen flow-matching VLA policies, while showing that critic generalization and uncertainty-aware guidance remain the central bottlenecks.

Value functions are an essential component in actor-critic based deep reinforcement learning (RL). Conventionally, these functions are trained as a regression task by minimising the mean squared error (MSE) relative to bootstrapped target values. Meanwhile, in distributional RL, a distribution of returns is modelled based on the distributional Bellman operator. This work investigates the Gaussian Histogram Loss (HL-Gauss), a recent approach that reframes value estimation as classification by encoding each scalar Bellman target as a Gaussian-smoothed categorical target. Despite its potential, applying histogram-based losses to RL presents inherent challenges, most notably the requirement to pre-define a fixed support interval, which is often complicated by the non-stationary and stochastic nature of target values typically found in RL tasks. In this work, we propose an approach that dynamically learns the lower and upper bounds of the support instead of assigning them beforehand. We derive an objective that jointly learns these bounds whilst learning the categorical representation of the scalar values, and we show that this objective forms an upper bound on the mean-squared Bellman error. Our theoretical analysis further shows that this bound is tighter than that of non-learned supports of HL-Gauss. Empirically, the proposed objective enables stable adaptation of the support interval and matches HL-Gauss-based actor-critic algorithms on most continuous-control tasks whilst improving on a subset, without requiring a pre-specified support interval.

Actor-critic methods depend on learned critics, but critic quality is often evaluated only indirectly through return, temporal-difference error, or value loss. Critic complexity is introduced as an additional diagnostic and intervention dimension for actor-critic reinforcement learning. The analysis uses spectral effective-rank entropy, a rank-like summary of the singular-value distributions of critic weight matrices, to assess critic model complexity. Across TD3 and PPO experiments, critic complexity is tracked together with return and Monte Carlo value-estimation bias. The results show that critic complexity is measurable throughout training and is systematically associated with training behavior, while also making clear that the relationship is heterogeneous across algorithms, tasks, and hyperparameters. A direct complexity-control intervention is then evaluated by adding a spectral-entropy penalty to the critic loss. This intervention reliably changes the targeted spectral quantity, demonstrating that critic complexity can be controlled rather than only observed. Return effects are treated as task-dependent evidence rather than as a general performance claim, because overall complexity-control results vary.