Reinforcement Learning

Process reward models (PRMs) have become a key component for LLMs, as their step-level feedback supports both post-training and test-time reasoning. However, training strong PRMs remains costly: human step annotation is difficult to scale, while Monte Carlo estimation is computationally expensive and can drift from the intrinsic correctness of steps. To get effective PRMs at low cost, we introduce TIPS (Thinking-Induced Process Supervision), an outcome-only reinforcement learning (RL) framework for training generative PRMs. In TIPS, the model generates a chain-of-thought (CoT) followed by step-level labels and an outcome label. The reward depends solely on whether the predicted outcome matches the ground truth, and the resulting group-relative advantage is used to optimize the entire generated response. Intuitively, when checking intermediate steps helps determine the outcome, more accurate checks can lead to better outcome judgments and higher rewards. Outcome-only RL can therefore reinforce step-level verification without explicit process supervision. We validate the effectiveness of TIPS across math and agent benchmarks and four backbone families. Notably, TIPS-Qwen3-4B-Thinking-2507 reaches 85.2 F1 on ProcessBench with only 3.2K outcome-labeled trajectories, surpassing all evaluated trained PRMs and strong prompt-only judges such as GPT-5.4-Instruct and Claude-4.7-Opus, while still trailing o1-mini. Code and data are available at https://github.com/RUCBM/TIPS.

Maximum Likelihood Reinforcement Learning (MaxRL) targets prompt-wise log-success and has shown strong performance on reasoning tasks. Under finite rollout budgets, however, the estimator used by MaxRL attenuates each prompt's likelihood gradient by a factor that depends on its success probability and rollout count. Under uniform rollout allocation, the common rollout count fails to compensate for success-dependent attenuation, leaving low-success prompts more strongly attenuated and distorting their relative contributions to the expected aggregate gradient. We introduce SERA (Scale-Equalized Rollout Allocation), which redistributes a fixed rollout budget to approximately equalize these finite-rollout scaling factors. Building on our theoretical analysis of how finite rollouts distort prompt-wise likelihood gradients, we formulate the allocation as a fixed-budget max--min problem, derive a waterline solution to its continuous relaxation, and introduce a multiplicity correction to remove the additional prompt weighting induced by heterogeneous rollout counts. Experiments show stronger alignment with exact likelihood gradients in a controlled ImageNet setting and improved multi-sample solution coverage over MaxRL on maze navigation and mathematical reasoning under matched training rollout budgets.

Agentic reinforcement learning (ARL) with verifiable rewards improves the ability of large language models (LLMs) to tackle knowledge-intensive tasks by learning to interleave search and reasoning. However, most existing ARL methods optimize only LLM-generated tokens and treat retrieved evidence as environment observations. This creates an information-credit gap: failures caused by missing or misleading evidence are attributed to the LLM policy rather than to the retriever, which motivates training the LLM and the retriever jointly. In this paper, we show that retrieval and LLM policy learning are order-sensitive: adapting the retriever before optimizing the policy yields a larger reward gain than the reverse order. To preserve this hierarchy while allowing both components to co-adapt, we formulate retrieval-augmented agentic RL as a bilevel optimization problem. To solve it efficiently, we introduce BRIDGE, a memory-efficient first-order bilevel method motivated by a loss-landscape analysis of the RL and retrieval objectives. Across seven open-domain QA benchmarks, BRIDGE achieves the highest average accuracy with both 3B and 7B backbones, improving the multi-hop average over the strongest baseline by 9.6 and 3.4 EM points, respectively. It also achieves the best averaged answer accuracy and reasoning quality across medical QA benchmarks.

Hardware accelerators based on physical dynamical systems offer an attractive route toward energy-efficient reinforcement learning applications. However, their scalability is challenging because it requires many statistically independent entropy sources. Here, we introduce a quasi-analog decision-making architecture based on asynchronous Boolean networks (or lattices) implemented on a clockless reconfigurable chip. Each node in the network consists of a single logic element that acts as an autonomous entropy source. This architecture gives rise to distributed Boolean chaos, in which a spatially coupled network generates parallel streams of chaotic Boolean transitions with very low statistical dependence between nodes. We experimentally demonstrate parallel decision-making on a 1024-armed bandit problem, which is beyond the scale of previous hardware implementations, while significantly improving power-law scaling performance. Separately, we scale the proposed entropy source to 5120 parallel channels, yielding an aggregate sample generation rate of 2.14 TS/s. Our solution is implemented on a commercial reconfigurable CMOS chip and offers high integration density and ease of programmability. Our results pave the way for using distributed Boolean chaos as a valuable hardware substrate for large-scale reinforcement learning and for the development of fully integrated, high-throughput decision-making accelerators.

A goal that is close in space can be far away in time. Obstacles, terrain, and the agent's own capabilities determine how long it takes to get there. Yet, critics in contrastive and survival reinforcement learning do not measure the distances in their representation space in units of time. We therefore introduce ChronoSRL, which gives the critic's embeddings an explicit temporal geometry. The distance between state-action and goal embeddings is trained to match the time that the agent takes to reach the goal (goal-reaching time), while goals that were not reached, and goals from other trajectories, are pushed at least one discount horizon away. Furthermore, reaching a goal quickly once does not mean that reaching it is reliable in general, so the policy should not follow the temporal distance directly. Instead, we build on survival reinforcement learning and predict from our temporal embeddings not only the full distribution of goal-reaching times but also the time spent near the goal. Thereby, the policy is trained to favor actions that reach the goal sooner and more reliably and that keep the agent near it. ChronoSRL learns faster and reaches higher performance than contrastive, action-chunked contrastive, and survival reinforcement learning baselines on seven standard locomotion and navigation benchmarks, even with much smaller networks. To test the limits of self-supervised reinforcement learning, we introduce velocity tracking, goal-position reaching, and box climbing tasks with a quadruped robot in a realistic sim-to-real locomotion setup, and show how the shaping terms that are typical for robotics can be naturally incorporated into our framework. ChronoSRL is the only one of the tested self-supervised reinforcement learning methods that learns to stay at the commanded velocities and goal positions, and climbs the highest boxes.

Reinforcement learning (RL) in simulation can train dexterous manipulation policies without robot demonstrations, but training a single generalist policy with task-agnostic rewards faces a severe exploration problem: approaching, grasping, and reorienting diverse objects with many degrees of freedom is difficult to discover from scratch. Prior works make exploration tractable with high-quality robot demonstrations, per-task reward shaping, or by restricting policies to narrow modes of behavior. We propose X-Reset, a framework that instead resolves exploration with human hand-object demonstrations. Rather than imitating or tracking retargeted human motion, X-Reset kinematically retargets hand-object states to noisy robot states, filters out states that are unstable in simulation, and samples the remainder as resets during RL training with general-purpose object-centric rewards. The resulting policy depends only on object state and goal, with demonstrations entering training through the reset distribution. We show that X-Reset trains generalist policies on 20 objects across three embodiments---a 22-DoF hand on two different arms and a parallel-jaw gripper---and resolves the exploration challenges of RL from scratch. X-Reset scales with the number of training objects, generalizes to unseen objects, can learn from imperfect hand-pose estimates, and transfers behaviors zero-shot from sim-to-real.

We study adversarial imitation learning (AIL), in which an agent learns to imitate expert demonstrations by optimizing a policy against an adversarial reward that distinguishes expert and learner behavior. Historically, reward regularization and entropy-based policy regularization are key components of empirically successful methods such as GAIL and LS-IQ, yet their finite-sample benefits remain underexplored. We establish fast rates for jointly regularized AIL in finite-horizon Markov decision processes with general function approximation. Our model-free algorithm, Dually Regularized AIL, combines KL policy regularization with a quadratic reward penalty weighted by expert and learner occupancies. With K online episodes and N expert trajectories, we prove a $\widetilde{O}\left(\frac{1}{K}+\frac{1}{N}\right)$ bound on the regularized imitation gap for fixed regularization parameters. Our analysis combines an online mirror descent construction for general convex reward classes to control estimation error from finite expert data and stochastic learner feedback, with a sharp analysis of optimistic KL-regularized policy learning. To the best of our knowledge, Dually Regularized AIL is the first algorithm to simultaneously achieve $\widetilde{O}\left(\frac{1}ε\right)$ sample complexity in both expert demonstrations and online interactions for this regularized AIL objective, even with stochastic experts. These results provide a rigorous characterization of the complementary statistical benefits of reward and policy regularization in AIL.

In reinforcement learning with verifiable rewards (RLVR), imperfect verifiers can reward incorrect responses, creating opportunities for reward hacking. Using gradient flow with a fixed verifier, we characterize the conditions under which reward rises while correctness falls. We then show that the observations available during RLVR are, in general, insufficient to detect or identify accepted errors, or to guarantee their reduction without sacrificing correct responses. To address this limit, we construct a correction using additional feedback about correctness from audits. This correction achieves \emph{selective control}: at the current policy, it lowers the probability of accepted errors and raises that of correct responses, provided it outweighs the pressure toward errors from verifier reward. Experiments with log linear and neural contextual bandits and with a language model support the analysis and show that selective control under partial auditing reduces accepted errors while increasing correctness.

Behavioral Foundation Models (BFMs) are an emerging paradigm in reinforcement learning, playing a role analogous to large language models in natural language processing: they have shown remarkable versatility, enabling zero-shot performance, fast imitation, and online adaptation, all by exploiting the structure of a latent space. In this work, we investigate whether the latent behavioral space induced by BFMs can serve as an effective search space to discover large repertoires of behaviorally diverse and high-performing policies through Quality-Diversity (QD) methods. While QD methods generally search directly in high-dimensional policy parameter space, in this paper, we present BFM-QD, a framework that performs QD search in the compact latent space of a BFM. We further show that the BFM-QD framework provides a closed-form, gradient-free policy improvement operator that approximates a policy gradient update, but requires no critic training and no backpropagation. Across continuous-control benchmarks spanning dense locomotion, sparse navigation, and contact-rich manipulation, BFM-QD consistently outperforms parameter-space baselines, with particularly stark gains in sparse and deceptive settings, where all tested parameter-space QD methods collapse to near-zero performance. These results show the effectiveness of the BFM-QD framework, benefiting from the synergy between dimensionality reduction of the search space and offline pretraining from diverse behavioral data. This positions BFMs as a general-purpose backbone for QD optimization, extending their utility beyond zero-shot task solving to the discovery of diverse behavioral repertoires.

Soft Actor-Critic (SAC) is widely used for entropy-regularised reinforcement learning with continuous action spaces, and practical implementations perform only a few actor steps towards an evolving target. In this work, we prove convergence guarantees when the target policy arises from policy mirror descent and compare it with the classical Gibbs target. We derive sufficient conditions for the strong convexity and smoothness of the actor objective, characterised by the curvature of the $Q$-function estimate through the Legendre differential operator, and establish an $\mathcal{O}\!\left(N^{-\frac{1}{5}}\right)$ best-iterate finite-time convergence rate up to actor and critic approximation errors. Moreover, the mirror-descent step size $λ$ directly controls the target drift and hence actor tracking error, whereas the analogous Gibbs bound contains a non-vanishing tracking term.

Learning energy-efficient humanoid locomotion requires discovering mechanically economical gait coordination, not merely reducing actuator effort. Reinforcement learning promotes efficiency through effort-related reward penalties, which guide the step-to-step mechanics of walking only indirectly. This article proposes a framework inspired by passive dynamic walking (PDW) that temporarily creates slope-equivalent conditions favorable to economical gait discovery and removes all PDW-specific guidance before nominal-dynamics optimization. During early training, a tilted-gravity field assists sagittal progression on flat collision geometry, complemented by curriculum-coupled reward terms. The core framework requires no reference trajectories, gait phases, or contact schedules. In a five-seed forward-locomotion study on a 29-DoF Unitree G1, the framework reduces mechanical cost of transport by 6.8-15.2% over commanded speeds of 0.5-2.0m/s without degrading velocity tracking. Mechanical-work decomposition attributes the reduction to positive actuator work, and reward-matched comparisons separate the guided regime's faster gait acquisition from the tilt's additional benefit to converged economy. The framework extends to unassisted omnidirectional locomotion, where its benefit persists once a walking-specific motion prior supplies kinematic coordination, the combination reducing speed-matched cost of transport by 18.7%. On hardware, forward cost of transport falls by 16.3% with the motion prior and by 4.5% without it, the latter within the trial-to-trial spread.

Group Relative Policy Optimization (GRPO) improves language-model reasoning by comparing verified rewards among multiple solution rollouts for each query. However, difficult training queries can yield only incorrect rollouts, leaving GRPO with no reward contrast or learning signal. Prior hint-based methods construct auxiliary hints from solution evidence and use them to re-solve failed queries, recovering learning signal. Yet the resulting trajectories are typically treated as ordinary solution trajectories despite being generated under an assisted condition unavailable at evaluation. We discover hinted reward shift: recovered reward contrast can concentrate policy updates on hinted trajectories, limiting improvement without hints. This also creates a trade-off: increasing hinted trajectories can accelerate early learning but intensify reward shift later. To address this problem, we propose HATCH (Hint-Annealed Self-Teaching), an online single-policy framework that learns from both generating and using its own hints to improve reasoning without assistance. To mitigate hinted reward shift, we introduce online weighting to anneal the contribution of hinted trajectories. However, learning to generate hints can conflict with improving query solving. We therefore use gradient projection to remove the opposing component of hint-generation updates. Together, these designs support self-improvement by enabling the policy to create learning opportunities for itself and turn them into stronger reasoning without hints. We evaluate our method on mathematical reasoning benchmarks and outperform state-of-the-art methods by 1.02 pp on Llama-3.2-1B-Instruct, 2.84 pp on Qwen3-1.7B, and 4.32 pp on Qwen3-8B.

Deep reinforcement learning (DRL) algorithms for movement control are typically evaluated and benchmarked on sequential decision tasks where imprecise actions may be corrected with later actions, thus allowing high returns with noisy actions. In contrast, we focus on an under-researched class of high-risk, high-precision motion control problems where actions carry irreversible outcomes, driving sharp peaks and ridges to plague the state-action reward landscape. Using computational pool as a representative example of such problems, we propose and evaluate State-Conditioned Shooting (SCOOT), a novel DRL algorithm that builds on advantage-weighted regression (AWR) with three key modifications: 1) Performing policy optimization only using elite samples, allowing the policy to better latch on to the rare high-reward action samples; 2) Utilizing a mixture-of-experts (MoE) policy, to allow switching between reward landscape modes depending on the state; 3) Adding a distance regularization term and a learning curriculum to encourage exploring diverse strategies before adapting to the most advantageous samples. We showcase our features' performance in learning physically-based billiard shots demonstrating high action precision and discovering multiple shot strategies for a given ball configuration.

Tree-structured reinforcement learning trains search agents by comparing alternative continuations and propagating terminal rewards to intermediate decisions. Adaptive expansion, however, creates a statistical asymmetry: an incumbent is selected using its own generation statistic, whereas fresh siblings are sampled after selection. When that statistic is associated with return, branch values can reflect selection history as well as continuation quality, even for a shared parent. We propose Selective-Inference Policy Optimization (\SIPO{}), which incorporates this distinction into tree-based credit estimation. Its scale-free branch criterion keeps generation scores and sibling penalties on a consistent relative scale; exchangeable branching supplies multiple fresh continuations from each selected parent; and order-statistic correction adjusts retained incumbent values using selection rank and the estimated score--outcome association. These mechanisms preserve the leaf budget and the host policy optimisation objective. Across seven QA benchmarks using Qwen3-4B, Qwen3-8B, and Qwen2.5-7B, \SIPO{} achieves the highest reported multi-hop and single-hop averages among the compared methods. On Qwen3-8B, it improves these averages over AT\textsuperscript{2}PO by $1.31$ and $1.07$ percentage points, respectively, and ranks first on six of seven benchmarks. Component ablations evaluate the individual and combined changes, while early-training paired diagnostics show a selected--fresh value gap alongside a near-zero fresh--fresh reference. Together, these results support accounting for selection history when constructing and evaluating search-agent rollouts. Our code is available at https://github.com/Zenghuang-Fu/SIPO

Reward-maximizing reinforcement learning (RL) is widely used to post-train stochastic diffusion and flow policies for text-to-image (T2I) generation. However, reward-maximizing RL causes policy mode collapse even under reference KL or entropy regularization, reducing the policy to a single high-reward mode. In T2I, this produces similar images and reward hacking. When extended to vision-language-action (VLA) models, the same collapse removes alternative successful strategies and weakens task and scene generalization. To address this limitation, we introduce Unified Trajectory Matching Policy Optimization (Uni-TMPO), a unified RL post-training framework for diffusion and flow policies. First, Uni-TMPO converts standardized rewards into a target distribution within each trajectory group and derives the policy distribution from trajectory log probabilities. Then, forward Kullback-Leibler optimization matches the two distributions instead of maximizing expected reward. A progress-conditioned coarse-to-fine scheduler efficiently constructs T2I trajectories. Within the unified framework, feedback-conditioned sampling uses updated observations to construct VLA trajectories. Extensive experiments show that Uni-TMPO achieves higher T2I rewards and VLA ID success rates than the strongest baselines. More importantly, it achieves the best T2I reward-diversity-efficiency trade-off and VLA generalization to held-out tasks and scenes, while real-robot evaluation demonstrates the value of multiple action strategies when the higher-reward target is blocked.

Looped architectures scale computation by reusing the same parameters across recurrent steps, and recent work shows that they substantially improve deep reinforcement learning policies on long-horizon tasks. Since recurrent depth directly controls computation, one may expect looped policies to naturally support elastic inference across recurrent depths. Surprisingly, we find that pretrained looped policies exhibit severe recurrent-depth specialization: reliable decisions are concentrated near the full trained depth, tying deployment computation to this depth even when less computation may suffice. Achieving depth elasticity, i.e., reliable decisions across recurrent depths with adaptive computation at deployment, therefore remains a key challenge. To address this, we propose FlexLoop, a novel post-training framework that converts pretrained fixed-depth looped policies into depth-elastic policies. FlexLoop keeps training on the original RL objective to preserve full-depth capability while performing adjacent-depth policy distillation to progressively transfer decision quality from deeper to shallower recurrent steps. The resulting policy supports reliable inference across recurrent depths and enables state-wise adaptive inference through recurrent-depth consistency. Experiments on $30$ online and offline long-horizon goal-conditioned environments show that FlexLoop preserves full-depth performance while making shallower depths effective. Keeping competitive performance, FlexLoop reduces average recurrent depth by up to $\bf{43\%}$ and achieves up to $\bf{1.34\times}$ wall-clock speedup in a stress test.

Efficient cooperation is challenging due to the usual partial observability of each agent in multi-agent reinforcement learning. Recurrent networks encode local interaction histories, but their hidden representations provide limited insight into the information underlying individual decisions. To address these challenges, we propose a novel interpretable framework, called escaping local views (ELV), which introduces semantically structured latent concepts to render policy decisions transparent. Specifically, each agent extracts low-dimensional semantic concepts from its local observation and action-observation trajectory. These concepts are jointly encoded into a contextual latent variable via a variational autoencoder (VAE), which builds a bridge between local views and global semantics. To explicitly model the decision of each agent, we employ a dual-path attention mechanism in which one module estimates the salience of individual concepts relative to the global context, while the other captures higher-order cooperative patterns with pairwise concept interactions. Furthermore, we incorporate a concept prediction module that derives an intrinsic reward from next-concept prediction errors, which incentivizes agents to explore regions of semantic novelty. Experiments in multiple environments verify that ELV not only achieves competitive performance but also explicitly provides how agents reason about their decisions.

Scaling LLM-based optimization from textbook-scale instances to real-world, industrial tasks remains a critical open challenge. Existing approaches are predominantly evaluated on small, self-contained textual problems and often commit to a solver-integrated paradigm, limiting their ability to handle the scale and structural diversity of practical optimization workloads. In this work, we propose a practical framework for training open-source LLMs to tackle real-world, industrial-scale optimization. We first show empirically that solver-integrated reasoning, exact combinatorial algorithm, and heuristic search exhibit complementary strengths across different problem structures and scales. Motivated by this, we introduce Strategy-Diverse Reinforcement Learning (SDRL), which trains LLMs as adaptive optimization meta-solvers. SDRL leverages this complementarity through a correctness-gated hierarchical diversity reward that promotes robust exploration across varying strategies and within each strategy, effectively preventing premature strategy collapse. We further introduce a mixed-format training scheme that jointly supports both self-contained textual problems and file-grounded instances. Across comprehensive evaluations, our framework outperforms existing fine-tuned methods and frontier models including DeepSeek-V4-Pro and GPT-5.5, both on average across benchmarks and on industrial-scale optimization tasks.

This paper addresses the problem of safe offline reinforcement learning, which involves training a policy to satisfy safety constraints using an offline dataset. This problem is inherently challenging as it requires balancing three highly interconnected and competing objectives: satisfying safety constraints, maximizing rewards, and adhering to the behavior regularization imposed by the offline dataset. To tackle this trilogy challenge, we propose Q-learning Penalized Transformer policy (QPT), a \emph{training--inference consistent} framework that bridges conditional sequence modeling with constraint-aware value estimation. QPT trains a Transformer policy that generates actions conditioned on trajectory context and target return/cost, retaining strong behavior regularization. To inject explicit safety semantics during learning, we augment sequence-model training with a Q-shaped penalty using learned reward and cost Q-functions to favor high return under low constraint violation. At inference, the same Q-functions enforce the cost threshold and choose the highest-reward feasible action, closing the loop between training and deployment. We provide a principled analysis under stylized near-deterministic CMDPs, characterizing how Q-penalized conditional generation improve safety and performance. Empirically, QPT consistently outperforms strong safe offline RL baselines across 38 tasks on the DSRL benchmark, and exhibits robust zero-shot adaptation to different constraint thresholds.

Glove-based motion capture is emerging as a scalable approach to collecting dexterous-hand demonstration data. However, due to the kinematic gap between the human and robot hand, the recorded human motions cannot be executed directly on the robot, especially for contact-rich tool-use tasks involving in-hand reorientation. Prior work bridges this gap in simulation through reinforcement learning (RL) or trajectory optimization, but the human contact pattern is hard to preserve under such formulations, often producing unnatural manipulation and unstable functional grasps. These methods also train a separate policy or solve a separate optimization for each reference trajectory, which is inefficient. To solve these problems, we propose DexTaG, a tactile-guided RL framework for dexterous manipulation. During training, tactile signals captured by the glove guide policy search toward the measured human contact pattern, reducing reliance on precise reference geometry for contact supervision. To improve efficiency, we train a single generalizable retargeter jointly on all training trajectories of the same object. The retargeter is further distilled into a tactile-free student controller conditioned on the target object trajectory for real-world deployment. On marker-pen and hammer manipulation tasks, DexTaG learns natural, contact-rich behaviors that baselines with distance-based contact heuristics fail to learn, generalizes to held-out trajectories of the same object and task, and outperforms single-trajectory baselines on OakInk2.

Large language model (LLM) agents increasingly undertake extreme-long (xlong) horizon tasks, where a single execution can span hours, hundreds of model--environment interactions, and nearly 1M tokens per rollout. Applying online reinforcement learning (RL) to such executions poses two fundamental challenges: (1) severe execution variance and prolonged rollout delays cause massive GPU idling; and (2) complex non-linear branching generates massive trajectory redundancy, crippling training efficiency. To address these, we presents QwenGyre, an end-to-end framework for xlong-horizon online RL. QwenGyre elastically reallocates GPUs between rollout and training without interrupting live executions, while its trajectory processor reconstructs branching histories, scores partial progress, and deduplicates redundant paths to bound training costs. Scaled to our flagship model, Qwen~3.8 2.4T, with 700K tokens per rollout, QwenGyre yields a 6.0% absolute gain on NL2RepoBench (52.5% $\to$ 58.5%) in 48 steps. Across our evaluations on diverse domains of training datasets, QwenGyre delivers up to $1.85\times$ and $1.78\times$ speedups over Colocate and Async, respectively.

Generative robot policies provide expressive behavior priors, but updating a large diffusion or flow-matching model through online interaction is costly. Latent-space reinforcement learning avoids updating the pretrained generator by controlling its initial sampling noise, yet high-dimensional noise can have strongly anisotropic effects on decoded actions. We introduce Principal Steering Subspaces (PSS), a forward-query interface that constructs a fixed low-dimensional control basis from finite-difference decoder responses. Soft Actor-Critic controls the leading response directions, while the orthogonal complement is independently resampled from the Gaussian prior at each query. On three RoboMimic tasks with diffusion and flow-matching policies, response spectra reveal substantial concentration. Across five matched task-generator pairs, the training curves indicate that PSS generally converges faster and exhibits more stable late-training behavior than full-latent control, while achieving stronger final performance overall. Controlled Diffusion-Square ablations further show that leading-response directions outperform random and least-responsive subspaces of equal dimension. We further integrate PSS with a frozen, closed-source 3B-parameter vision-language-action (VLA) policy in a humanoid learning system with synchronous transition collection, reset-time optimization, and latency-aware asynchronous deployment. In an exploratory screwdriver-placement evaluation, success is observed in 2/10 trials for the frozen VLA policy and 6/10 after SAC+PSS adaptation. These results support decoder-response geometry as a practical basis for online adaptation of frozen generative robot policies.

Speech-to-speech translation (S2ST) in time-sensitive applications such as video dubbing requires not only semantic fidelity and speaker preservation, but also strict duration consistency to avoid audio-visual misalignment. However, existing S2ST systems largely generate target speech without explicit temporal planning, making duration control an unresolved challenge. We introduce DuraS2ST, a duration-aligned reasoning framework that enables a single speech language model to first generate an explicit chain-of-thought (CoT) for planning target wording and phonetic length, and then synthesize the corresponding speech tokens. To support this paradigm, we construct DuraSet-440K, a high-quality duration-aligned CoT corpus for supervised initialization. We further optimize the model with multi-modal multi-dimensional reinforcement learning, using a Duration Margin Reward to balance translation quality and duration consistency, and Modality-Aware Reward Attribution to assign rewards to appropriate token spans. Experiments on CVSS-T show that DuraS2ST achieves a strong balance between translation quality and duration consistency, outperforming competitive open-source and commercial baselines. Project page: https://github.com/Mia11939/DuraS2ST.

A nascent family of methods that forgoes the policy gradient and reweights a supervised regression instead has garnered momentum in reinforcement learning for diffusion and flow models. DiffusionNFT, FlowAWR, and RAM are representative regimes with contrasting motivations. It is yet opaque what, if anything, they share. We substantiate that each is the solution of one divergence-constrained reward-maximization problem, and they are differentiated only by the convex generator that defines the constraint. Under the unified modeling framework, we unravel the relaxations that prior art made during building the advantage-embedded regression target: approximating the KKT condition and posterior normalizer for the linear and exponential tilt shapes DiffusionNFT and FlowAWR respectively, while preserving the exact sparsemax projection onto the probability simplex for linear tilt leads to another superior model type in this work. Beyond the theoretical underpinnings, we further empirically investigate the design space and shed light on the training recipe for regression-style diffusion RL. Retaining the merits discovered during our exploration gives rise to DiffusionRFT, our paradigm that converges faster, trains more stably, and attains the top performance.

In offline goal-conditioned reinforcement learning (GCRL), divide-and-conquer scales to long horizons by joining two shorter segments at a subgoal. However, under stochastic dynamics, the base case of this rule values the luckiest trajectories through the data. The subgoal must also lie on a shared trajectory, so a state-goal pair that no trajectory connects gets no value update at all. To address both, we present Grounded Transitive RL (GTRL), an offline GCRL value learning algorithm that grounds the divide-and-conquer update with a one-step TD target. Over a single step, TD is correct, as its target averages over the successors and needs no subgoal. GTRL adds this target to the composition rather than replacing it, so every pair receives an update, and the composition still carries the long horizon. GTRL also corrects the bias from hindsight relabeling by reweighting each goal against how reachable it was from other successors. We evaluate our algorithm on nineteen OGBench tasks spanning stochastic, deterministic, and stitching environments, where it achieves the highest average success rate. Code will be released soon.

Query understanding (QU) plays a critical role in production search systems, translating raw user queries into search execution plans that drive downstream retrieval and ranking. While large language models (LLMs) have enabled QU to be framed as a structured multi-task generation problem (e.g., intent classification, query expansion), optimizing such models to produce search-engine-coupled outputs remains challenging: static, label-based supervision fails to capture how each component actually interacts with the underlying search pipeline to affect downstream performance. We present a search-aware reinforcement learning (RL) framework for QU based on a distill-then-RL paradigm. Teacher-student supervised fine-tuning (SFT) first yields a well-formed, schema-compliant policy initialization. The RL stage then optimizes each QU component with rewards derived from live interaction with the search engine, tailored to that component's operational role, rather than a single reward tied to the final search outcome. Experiments on Roblox search show that this component-specific optimization improves both per-component utility and downstream search quality, raising NDCG@20 by 8.9 points over the SFT policy and by 3.5 points over training with a single end-to-end reward.

Detectors of alignment failures screen deployed language models and score alignment benchmarks. Most are generative judges that spend a decoding pass on every criterion, and classifiers that read token probabilities, such as Llama Guard, still score one fixed label per call. Jev, a model trained with reinforcement learning for calibrated decisions (RLCD), answers many typed questions about one input with calibrated probabilities in a single call. Whether it detects alignment failures has not been measured. We present RLCDAlignBench, which benchmarks Jev on ten alignment failures: sycophancy, jailbreaks, deception, prompt injection, hallucination, privacy violation, social bias, reward hacking, concealing uncertainty, and power seeking. It spans 44 benchmarks and five target models, labelled by each benchmark's scorer and, on two, by humans. Many of these failures are relational, defined against a reference, such as the user's belief or an injected instruction, that the response alone does not reveal. Our key idea is therefore to vary what Jev is asked separately from what it sees: the question's wording and answer type on one side, the fields of the input on the other. A single generic question reaches a median AUROC of 0.886 zero-shot and beats supervised baselines on most benchmarks. Question wording matters little, while context matters more, mostly through fields that encode the label. Jev matches the reference scorer's agreement with human labels, surfaces label defects in existing benchmarks, and costs 63x less than LLM-judge scorers. Code and data: https://github.com/sumleo/RLCDAlignBench.

There are many complex issues in the world of artificial intelligence. Some of these problems are solved using other artificial intelligence methods, which are called artificial intelligence for artificial intelligence. Finding an appropriate classifier algorithm is a time-consuming task. For this reason, an algorithm that can automatically learn the choice of classification algorithms is very important. Classification algorithms are useful in predicting various diseases. Also, Primary Biliary Cirrhosis is one of the most well-known diseases that have been predicted by classification algorithms. This research's most significant achievement and novelty is the automatic increase in learning through a scoring method of reinforcement learning is called square learning (SL). In this research, an algorithm is presented that learns to automatically select the appropriate classification algorithm to predict Primary Biliary Cirrhosis. In this article, with inspiration from four evaluation metrics in classification algorithms, a new reinforcement learning method by the name of Fourth Degree Learning has been presented. In this research, we increased the performance of the classification algorithms used in this method from 63% of accuracy and achieved 98% accuracy.

Over the course of a lifetime, robots may encounter novel scenarios unaccounted for in its original training that result in performance degradation. One common approach to mitigating this issue is to further grow the offline training dataset in hopes of producing a policy robust to these changes. In contrast, biological learning occurs moment-to-moment via a stream of experience, unlike the predominantly batch-based and offline nature of deep learning. Although recent works show the feasibility of stream-based deep reinforcement learning, where updates use only the latest experience, none have shown it to be a viable continual learning framework for adapting robotic policies to unseen changes. In this paper, we present the first analysis of streaming deep reinforcement learning for adaptive continual learning in robotics. In particular, we show that, following an initial pretraining phase, streaming deep RL can enable a robot to successfully adapt to unforeseen changes to itself, its environment, or goals. Our primary experiments within quadruped locomotion demonstrate that a deep neural network robotic policy with certain optimizers and plasticity loss mitigation techniques can successfully leverage domain task knowledge from its pretraining to quickly adapt online to diverse changes via stream learning, outperforming batch-based on-policy methods and improving task success rates by up to 90% over the pretrained policy. Furthermore, we perform additional evaluations on robotic manipulation tasks to determine if our previous observations extend to different robotic morphologies and scenarios. Our results show that the successes observed in quadruped locomotion can be partially realized in manipulation with stability and performance limitations. We conclude with a discussion on the limitations of our work and its implications for the future of continual robot learning.

Reinforcement Learning with Verifiable Rewards (RLVR) performs well on problems with clear rewards, such as mathematics and coding, but whether it also works where the reward is less clear remains open. The reason-over-search recipe applies RLVR to open-domain question answering, where retrieval grounds the answer and a match against the reference supplies the reward. So far it has been demonstrated on large models, and below one billion parameters only with distillation from a larger teacher. We test the recipe on a small model. We train Qwen3.5-0.8B with Group Relative Policy Optimization (GRPO) and an interleaved Wikipedia-search tool on MuSiQue, varying only the reward across three shapes over three seeds each, and we evaluate every checkpoint held-out on a seven-benchmark question-answering suite. The recipe works: the best run reaches 0.352 average exact match against a 0.092 untrained floor, a 3.8-fold gain, with no distillation step in the training loop. The reward shape also matters. The Search-R1-faithful exact-match-only reward is the worst of the three at every seed at the matched training horizon, and it is worst even on exact match, the metric it directly optimises. We conclude that the sparse exact-match reward, RLVR's default in mathematics and code, is the wrong starting point for models of this size. The reason-over-search setting can supply a suitable reward for RLVR on small models, but small-model RLVR needs its own reward-design study rather than a scaled-down copy of a large-model recipe.