The T-maze, shown below, is a prototypical example of a task studied in the field of reinforcement learning. An artificial agent enters the maze from the left and immediately receives one of two possible observations: red or green. Red means that the agent will be rewarded for moving to the top at the right end of the corridor (in the question mark tile), while green means the opposite: the agent will be rewarded for moving down. While this seems like a trivial task, modern machine learning algorithms (such as Q-learning) fail at learning the desired behavior. This is because these algorithms are designed to solve (MDPs).

We consider the problem of learning the optimal policy in a discounted, infinite-horizon reinforcement learning (RL) setting where the reward signal is subject to adversarial corruption. Such corruption, which may arise from extreme noise, sensor faults, or malicious attacks, can severely degrade the performance of classical algorithms such as Q-learning. To address this challenge, we propose a new provably robust variant of the Q-learning algorithm that operates effectively even when a fraction of the observed rewards are arbitrarily perturbed by an adversary. Under the asynchronous sampling model with time-correlated data, we establish that despite adversarial corruption, the finite-time convergence rate of our algorithm matches that of existing results for the non-adversarial case, up to an additive term proportional to the fraction of corrupted samples. Moreover, we derive an information-theoretic lower bound revealing that the additive corruption term in our upper bounds is unavoidable. Next, we propose a variant of our algorithm that requires no prior knowledge of the statistics of the true reward distributions. The analysis of this setting is particularly challenging and is enabled by carefully exploiting a refined Azuma-Hoeffding inequality for almost-martingales, a technical tool that might be of independent interest. Collectively, our contributions provide the first finite-time robustness guarantees for asynchronous Q-learning, bridging a significant gap in robust RL.

In robotic systems, the performance of reinforcement learning depends on the rationality of predefined reward functions. However, manually designed reward functions often lead to policy failures due to inaccuracies. Inverse Reinforcement Learning (IRL) addresses this problem by inferring implicit reward functions from expert demonstrations. Nevertheless, existing methods rely heavily on large amounts of expert demonstrations to accurately recover the reward function. The high cost of collecting expert demonstrations in robotic applications, particularly in multi-robot systems, severely hinders the practical deployment of IRL. Consequently, improving sample efficiency has emerged as a critical challenge in multi-agent inverse reinforcement learning (MIRL). Inspired by the symmetry inherent in multi-agent systems, this work theoretically demonstrates that leveraging symmetry enables the recovery of more accurate reward functions. Building upon this insight, we propose a universal framework that integrates symmetry into existing multi-agent adversarial IRL algorithms, thereby significantly enhancing sample efficiency. Experimental results from multiple challenging tasks have demonstrated the effectiveness of this framework. Further validation in physical multi-robot systems has shown the practicality of our method.

The ``Last Mile Challenge'' has long been considered an important, yet unsolved, challenge for autonomous vehicles, public service robots, and delivery robots. A central issue in this challenge is the ability of robots to navigate constrained and cluttered environments that have high agency (e.g., doorways, hallways, corridor intersections), often while competing for space with other robots and humans. We refer to these environments as ``Social Mini-Games'' (SMGs). Traditional navigation approaches designed for MRN do not perform well in SMGs, which has led to focused research on dedicated SMG solvers. However, publications on SMG navigation research make different assumptions (on centralized versus decentralized, observability, communication, cooperation, etc.), and have different objective functions (safety versus liveness). These assumptions and objectives are sometimes implicitly assumed or described informally. This makes it difficult to establish appropriate baselines for comparison in research papers, as well as making it difficult for practitioners to find the papers relevant to their concrete application. Such ad-hoc representation of the field also presents a barrier to new researchers wanting to start research in this area. SMG navigation research requires its own taxonomy, definitions, and evaluation protocols to guide effective research moving forward. This survey is the first to catalog SMG solvers using a well-defined and unified taxonomy and to classify existing methods accordingly. It also discusses the essential properties of SMG solvers, defines what SMGs are and how they appear in practice, outlines how to evaluate SMG solvers, and highlights the differences between SMG solvers and general navigation systems. The survey concludes with an overview of future directions and open challenges in the field. Our project is open-sourced at https://socialminigames.github.io/.

Precise, correct feedback is crucial for effectively training large language models (LLMs) in code reinforcement learning. However, synthesizing high-quality test cases remains a profoundly challenging and unsolved problem. In this work, we present Klear-CodeTest, a comprehensive test case synthesis framework featuring rigorous verification to ensure quality and reliability of test cases. Our approach achieves broad coverage of programming problems via a novel Generator-Validation (G-V) framework, ensuring correctness through a consistency validation mechanism that verifies outputs against gold solutions. The proposed G-V framework generates comprehensive test cases including both regular and corner cases, enhancing test coverage and discriminative power for solution correctness assessment in code reinforcement learning. In addition, we design a multi-layered security sandbox system optimized for online verification platforms, guaranteeing safe and reliable code execution. Through comprehensive experiments, we demonstrate the effectiveness of our curated dataset, showing significant improvements in model performance and training stability. The source codes, curated dataset and sandbox system are available at: https://github.com/Kwai-Klear/CodeTest.

Congenital uterine anomalies (CUAs) can lead to infertility, miscarriage, preterm birth, and an increased risk of pregnancy complications. Compared to traditional 2D ultrasound (US), 3D US can reconstruct the coronal plane, providing a clear visualization of the uterine morphology for assessing CUAs accurately. In this paper, we propose an intelligent system for simultaneous automated plane localization and CUA diagnosis. Our highlights are: 1) we develop a denoising diffusion model with local (plane) and global (volume/text) guidance, using an adaptive weighting strategy to optimize attention allocation to different conditions; 2) we introduce a reinforcement learning-based framework with unsupervised rewards to extract the key slice summary from redundant sequences, fully integrating information across multiple planes to reduce learning difficulty; 3) we provide text-driven uncertainty modeling for coarse prediction, and leverage it to adjust the classification probability for overall performance improvement.

Recently, supervised learning (SL) methodology has emerged as an effective approach for offline reinforcement learning (RL) due to their simplicity, stability, and efficiency. However, recent studies show that SL methods lack the trajectory stitching capability, typically associated with temporal difference (TD)-based approaches. A question naturally surfaces: \textit{How can we endow SL methods with stitching capability and close its performance gap with TD learning?} To answer this question, we introduce $Q$-conditioned maximization supervised learning for offline goal-conditioned RL, which enhances SL with the stitching capability through $Q$-conditioned policy and $Q$-conditioned maximization. Concretely, we propose \textbf{G}oal-\textbf{C}onditioned \textbf{\textit{Rein}}forced \textbf{S}upervised \textbf{L}earning (\textbf{GC\textit{Rein}SL}), which consists of (1) estimating the $Q$-function by Normalizing Flows from the offline dataset and (2) finding the maximum $Q$-value within the data support by integrating $Q$-function maximization with Expectile Regression. In inference time, our policy chooses optimal actions based on such a maximum $Q$-value. Experimental results from stitching evaluations on offline RL datasets demonstrate that our method outperforms prior SL approaches with stitching capabilities and goal data augmentation techniques.

Hybrid storage systems (HSS) integrate multiple storage devices with diverse characteristics to deliver high performance and capacity at low cost. The performance of an HSS highly depends on the effectiveness of two key policies: (1) the data-placement policy, which determines the best-fit storage device for incoming data, and (2) the data-migration policy, which dynamically rearranges stored data (i.e., prefetches hot data and evicts cold data) across the devices to sustain high HSS performance. Prior works optimize either data placement or data migration in isolation, which leads to suboptimal HSS performance. Unfortunately, no prior work tries to optimize both policies together. Our goal is to design a holistic data-management technique that optimizes both data-placement and data-migration policies to fully exploit the potential of an HSS, and thus significantly improve system performance. We propose Harmonia, a multi-agent reinforcement learning (RL)-based data-management technique that employs two lightweight autonomous RL agents, a data-placement agent and a data-migration agent, that adapt their policies for the current workload and HSS configuration while coordinating with each other to improve overall HSS performance. We evaluate Harmonia on real HSS configurations with up to four heterogeneous storage devices and seventeen data-intensive workloads. On performance-optimized (cost-optimized) HSS with two storage devices, Harmonia outperforms the best-performing prior approach by 49.5% (31.7%) on average. On an HSS with three (four) devices, Harmonia outperforms the best-performing prior work by 37.0% (42.0%) on average. Harmonia's performance benefits come with low latency (240ns for inference) and storage overheads (206 KiB in DRAM for both RL agents combined). We will open-source Harmonia's implementation to aid future research on HSS.

Reinforcement Learning with Verifiable Rewards (RLVR) is a powerful paradigm for enhancing the reasoning ability of Large Language Models (LLMs). Yet current RLVR methods often explore poorly, leading to premature convergence and entropy collapse. To address this challenge, we introduce Curiosity-Driven Exploration (CDE), a framework that leverages the model's own intrinsic sense of curiosity to guide exploration. We formalize curiosity with signals from both the actor and the critic: for the actor, we use perplexity over its generated response, and for the critic, we use the variance of value estimates from a multi-head architecture. Both signals serve as an exploration bonus within the RLVR framework to guide the model. Our theoretical analysis shows that the actor-wise bonus inherently penalizes overconfident errors and promotes diversity among correct responses; moreover, we connect the critic-wise bonus to the well-established count-based exploration bonus in RL. Empirically, our method achieves an approximate +3 point improvement over standard RLVR using GRPO/PPO on AIME benchmarks. Further analysis identifies a calibration collapse mechanism within RLVR, shedding light on common LLM failure modes.

Vision-Language-Action (VLA) models have recently emerged as a powerful paradigm for robotic manipulation. Despite substantial progress enabled by large-scale pretraining and supervised fine-tuning (SFT), these models face two fundamental challenges: (i) the scarcity and high cost of large-scale human-operated robotic trajectories required for SFT scaling, and (ii) limited generalization to tasks involving distribution shift. Recent breakthroughs in Large Reasoning Models (LRMs) demonstrate that reinforcement learning (RL) can dramatically enhance step-by-step reasoning capabilities, raising a natural question: Can RL similarly improve the long-horizon step-by-step action planning of VLA? In this work, we introduce SimpleVLA-RL, an efficient RL framework tailored for VLA models. Building upon veRL, we introduce VLA-specific trajectory sampling, scalable parallelization, multi-environment rendering, and optimized loss computation. When applied to OpenVLA-OFT, SimpleVLA-RL achieves SoTA performance on LIBERO and even outperforms $π_0$ on RoboTwin 1.0\&2.0 with the exploration-enhancing strategies we introduce. SimpleVLA-RL not only reduces dependence on large-scale data and enables robust generalization, but also remarkably surpasses SFT in real-world tasks. Moreover, we identify a novel phenomenon ``pushcut'' during RL training, wherein the policy discovers previously unseen patterns beyond those seen in the previous training process. Github: https://github.com/PRIME-RL/SimpleVLA-RL