Self-improving systems require environmental interaction for continuous adaptation. We introduce SPICE (Self-Play In Corpus Environments), a reinforcement learning framework where a single model acts in two roles: a Challenger that mines documents from a large corpus to generate diverse reasoning tasks, and a Reasoner that solves them. Through adversarial dynamics, the Challenger creates an automatic curriculum at the frontier of the Reasoner's capability, while corpus grounding provides the rich, near-inexhaustible external signal necessary for sustained improvement. Unlike existing ungrounded self-play methods that offer more limited benefits, SPICE achieves consistent gains across mathematical (+8.9%) and general reasoning (+9.8%) benchmarks on multiple model families. Our analysis reveals how document grounding is a key ingredient in SPICE to continuously generate its own increasingly challenging goals and achieve them, enabling sustained self-improvement.

Site-specific disease management (SSDM) in crops has advanced rapidly through machine and deep learning (ML and DL) for real-time computer vision. Research evolved from handcrafted feature extraction to large-scale automated feature learning. With foundation models (FMs), crop disease datasets are now processed in fundamentally new ways. Unlike traditional neural networks, FMs integrate visual and textual data, interpret symptoms in text, reason about symptom-management relationships, and support interactive QA for growers and educators. Adaptive and imitation learning in robotics further enables field-based disease management. This review screened approx. 40 articles on FM applications for SSDM, focusing on large-language models (LLMs) and vision-language models (VLMs), and discussing their role in adaptive learning (AL), reinforcement learning (RL), and digital twin frameworks for targeted spraying. Key findings: (a) FMs are gaining traction with surging literature in 2023-24; (b) VLMs outpace LLMs, with a 5-10x increase in publications; (c) RL and AL are still nascent for smart spraying; (d) digital twins with RL can simulate targeted spraying virtually; (e) addressing the sim-to-real gap is critical for real-world deployment; (f) human-robot collaboration remains limited, especially in human-in-the-loop approaches where robots detect early symptoms and humans validate uncertain cases; (g) multi-modal FMs with real-time feedback will drive next-gen SSDM. For updates, resources, and contributions, visit, https://github.com/nitin-dominic/AgriPathogenDatabase, to submit papers, code, or datasets.

Abstract-- This paper presents a curriculum-based reinforcement learning framework for training precise and high-performance jumping policies for the robot'Jumper'. Separate policies are developed for vertical and horizontal jumps, leveraging a simple yet effective strategy. Next, a reference state initialization scheme is employed to accelerate the exploration of dynamic jumping behaviors without reliance on reference trajectories. We also present a walking policy that, when combined with the jumping policies, unlocks versatile and dynamic locomotion capabilities. Comprehensive testing validates walking on varied terrain surfaces and jumping performance that exceeds previous works, effectively crossing the Sim2Real gap. Experimental validation demonstrates horizontal jumps up to 1.25 m with centimeter accuracy and vertical jumps up to 1.0 m. Additionally, we show that with only minor modifications, the proposed method can be used to learn omnidirectional jumping. I. INTRODUCTION Quadruped robots can navigate complex terrains and overcome obstacles not only through walking but also through powerful jumps. The combination of robust walking and precise jumping capabilities is particularly valuable for planetary exploration [1], [2].

Reinforcement learning algorithms are typically designed for discrete-time dynamics, even though the underlying real-world control systems are often continuous in time. In this paper, we study the problem of continuous-time reinforcement learning, where the unknown system dynamics are represented using nonlinear ordinary differential equations (ODEs). We leverage probabilistic models, such as Gaussian processes and Bayesian neural networks, to learn an uncertainty-aware model of the underlying ODE. Our algorithm, COMBRL, greedily maximizes a weighted sum of the extrinsic reward and model epistemic uncertainty. This yields a scalable and sample-efficient approach to continuous-time model-based RL. We show that COMBRL achieves sublinear regret in the reward-driven setting, and in the unsupervised RL setting (i.e., without extrinsic rewards), we provide a sample complexity bound. In our experiments, we evaluate COMBRL in both standard and unsupervised RL settings and demonstrate that it scales better, is more sample-efficient than prior methods, and outperforms baselines across several deep RL tasks.

Neuromorphic computing systems are set to revolutionize energy-constrained robotics by achieving orders-of-magnitude efficiency gains, while enabling native temporal processing. Spiking Neural Networks (SNNs) represent a promising algorithmic approach for these systems, yet their application to complex control tasks faces two critical challenges: (1) the non-differentiable nature of spiking neurons necessitates surrogate gradients with unclear optimization properties, and (2) the stateful dynamics of SNNs require training on sequences, which in reinforcement learning (RL) is hindered by limited sequence lengths during early training, preventing the network from bridging its warm-up period. We address these challenges by systematically analyzing surrogate gradient slope settings, showing that shallower slopes increase gradient magnitude in deeper layers but reduce alignment with true gradients. In supervised learning, we find no clear preference for fixed or scheduled slopes. The effect is much more pronounced in RL settings, where shallower slopes or scheduled slopes lead to a 2.1x improvement in both training and final deployed performance. Next, we propose a novel training approach that leverages a privileged guiding policy to bootstrap the learning process, while still exploiting online environment interactions with the spiking policy. Combining our method with an adaptive slope schedule for a real-world drone position control task, we achieve an average return of 400 points, substantially outperforming prior techniques, including Behavioral Cloning and TD3BC, which achieve at most --200 points under the same conditions. This work advances both the theoretical understanding of surrogate gradient learning in SNNs and practical training methodologies for neuromorphic controllers demonstrated in real-world robotic systems.

Reinforcement learning (RL) in sparse-reward environments remains a significant challenge due to the lack of informative feedback. We propose a simple yet effective method that uses a small number of successful demonstrations to initialize the value function of an RL agent. By precomputing value estimates from offline demonstrations and using them as targets for early learning, our approach provides the agent with a useful prior over promising actions. The agent then refines these estimates through standard online interaction. This hybrid offline-to-online paradigm significantly reduces the exploration burden and improves sample efficiency in sparse-reward settings. Experiments on benchmark tasks demonstrate that our method accelerates convergence and outperforms standard baselines, even with minimal or suboptimal demonstration data.

Emergent symbolic representations are critical for enabling developmental learning agents to plan and generalize across tasks. In this work, we investigate whether large language models (LLMs) can translate human natural language instructions into the internal symbolic representations that emerge during hierarchical reinforcement learning. We apply a structured evaluation framework to measure the translation performance of commonly seen LLMs -- GPT, Claude, Deepseek and Grok -- across different internal symbolic partitions generated by a hierarchical reinforcement learning algorithm in the Ant Maze and Ant Fall environments. Our findings reveal that although LLMs demonstrate some ability to translate natural language into a symbolic representation of the environment dynamics, their performance is highly sensitive to partition granularity and task complexity. The results expose limitations in current LLMs capacity for representation alignment, highlighting the need for further research on robust alignment between language and internal agent representations.

Behavioral cloning is a simple yet effective technique for learning sequential decision-making from demonstrations. Recently, it has gained prominence as the core of foundation models for the physical world, where achieving generalization requires countless demonstrations of a multitude of tasks. Typically, a human expert with full information on the task demonstrates a (nearly) optimal behavior. In this paper, we propose to hide some of the task's information from the demonstrator. This ``blindfolded'' expert is compelled to employ non-trivial exploration to solve the task. We show that cloning the blindfolded expert generalizes better to unseen tasks than its fully-informed counterpart. We conduct experiments of real-world robot peg insertion tasks with (limited) human demonstrations, alongside videogames from the Procgen benchmark. Additionally, we support our findings with theoretical analysis, which confirms that the generalization error scales with $\sqrt{I/m}$, where $I$ measures the amount of task information available to the demonstrator, and $m$ is the number of demonstrated tasks. Both theory and practice indicate that cloning blindfolded experts generalizes better with fewer demonstrated tasks. Project page with videos and code: https://sites.google.com/view/blindfoldedexperts/home

End-to-end autonomous driving maps raw sensor inputs directly into ego-vehicle trajectories to avoid cascading errors from perception modules and to leverage rich semantic cues. Existing frameworks largely rely on Imitation Learning (IL), which can be limited by sub-optimal expert demonstrations and covariate shift during deployment. On the other hand, Reinforcement Learning (RL) has recently shown potential in scaling up with simulations, but is typically confined to low-dimensional symbolic inputs (e.g. 3D objects and maps), falling short of full end-to-end learning from raw sensor data. We introduce ZTRS (Zero-Imitation End-to-End Autonomous Driving with Trajectory Scoring), a framework that combines the strengths of both worlds: sensor inputs without losing information and RL training for robust planning. To the best of our knowledge, ZTRS is the first framework that eliminates IL entirely by only learning from rewards while operating directly on high-dimensional sensor data. ZTRS utilizes offline reinforcement learning with our proposed Exhaustive Policy Optimization (EPO), a variant of policy gradient tailored for enumerable actions and rewards. ZTRS demonstrates strong performance across three benchmarks: Navtest (generic real-world open-loop planning), Navhard (open-loop planning in challenging real-world and synthetic scenarios), and HUGSIM (simulated closed-loop driving). Specifically, ZTRS achieves the state-of-the-art result on Navhard and outperforms IL-based baselines on HUGSIM. Code will be available at https://github.com/woxihuanjiangguo/ZTRS.

Wheeled-legged robots combine the efficiency of wheels with the obstacle negotiation of legs, yet many state-of-the-art systems rely on costly actuators and sensors, and fall-recovery is seldom integrated, especially for wheeled-legged morphologies. This work presents a recovery-aware visual-inertial navigation system on a low-cost wheeled quadruped. The proposed system leverages vision-based perception from a depth camera and deep reinforcement learning policies for robust locomotion and autonomous recovery from falls across diverse terrains. Simulation experiments show agile mobility with low-torque actuators over irregular terrain and reliably recover from external perturbations and self-induced failures. We further show goal directed navigation in structured indoor spaces with low-cost perception. Overall, this approach lowers the barrier to deploying autonomous navigation and robust locomotion policies in budget-constrained robotic platforms.