Graphical user interface (GUI) agents have advanced rapidly but still struggle with complex tasks involving novel UI elements, long-horizon actions, and personalized trajectories. In this work, we introduce Instruction Agent, a GUI agent that leverages expert demonstrations to solve such tasks, enabling completion of otherwise difficult workflows. Given a single demonstration, the agent extracts step-by-step instructions and executes them by strictly following the trajectory intended by the user, which avoids making mistakes during execution. The agent leverages the verifier and backtracker modules further to improve robustness. Both modules are critical to understand the current outcome from each action and handle unexpected interruptions(such as pop-up windows) during execution. Our experiments show that Instruction Agent achieves a 60% success rate on a set of tasks in OSWorld that all top-ranked agents failed to complete. The Instruction Agent offers a practical and extensible framework, bridging the gap between current GUI agents and reliable real-world GUI task automation.

In supply chain management, decision-making often involves balancing multiple conflicting objectives, such as cost reduction, service level improvement, and environmental sustainability. Traditional multi-objective optimization methods, such as linear programming and evolutionary algorithms, struggle to adapt in real-time to the dynamic nature of supply chains. In this paper, we propose an approach that combines Reinforcement Learning (RL) and Multi-Objective Evolutionary Algorithms (MOEAs) to address these challenges for dynamic multi-objective optimization under uncertainty. Our method leverages MOEAs to search the parameter space of policy neural networks, generating a Pareto front of policies. This provides decision-makers with a diverse population of policies that can be dynamically switched based on the current system objectives, ensuring flexibility and adaptability in real-time decision-making. We also introduce Conditional Value-at-Risk (CVaR) to incorporate risk-sensitive decision-making, enhancing resilience in uncertain environments. We demonstrate the effectiveness of our approach through case studies, showcasing its ability to respond to supply chain dynamics and outperforming state-of-the-art methods in an inventory management case study. The proposed strategy not only improves decision-making efficiency but also offers a more robust framework for managing uncertainty and optimizing performance in supply chains.

Legged robots must achieve both robust locomotion and energy efficiency to be practical in real-world environments. Yet controllers trained in simulation often fail to transfer reliably, and most existing approaches neglect actuator-specific energy losses or depend on complex, hand-tuned reward formulations. We propose a framework that integrates sim-to-real reinforcement learning with a physics-grounded energy model for permanent magnet synchronous motors. The framework requires a minimal parameter set to capture the simulation-to-reality gap and employs a compact four-term reward with a first-principle-based energetic loss formulation that balances electrical and mechanical dissipation. We evaluate and validate the approach through a bottom-up dynamic parameter identification study, spanning actuators, full-robot in-air trajectories and on-ground locomotion. The framework is tested on three primary platforms and deployed on ten additional robots, demonstrating reliable policy transfer without randomization of dynamic parameters. Our method improves energetic efficiency over state-of-the-art methods, achieving a 32 percent reduction in the full Cost of Transport of ANYmal (value 1.27). All code, models, and datasets will be released.

The rapid expansion of scientific literature makes it increasingly difficult to acquire new knowledge, particularly in specialized domains where reasoning is complex, full-text access is restricted, and target references are sparse among a large set of candidates. We present a Deep Reinforcement Learning framework for sparse reference selection that emulates human knowledge construction, prioritizing which papers to read under limited time and cost. Evaluated on drug--gene relation discovery with access restricted to titles and abstracts, our approach demonstrates that both humans and machines can construct knowledge effectively from partial information.

Data collection is crucial for learning robust world models in model-based reinforcement learning. The most prevalent strategies are to actively collect trajectories by interacting with the environment during online training or training on offline datasets. At first glance, the nature of learning task-agnostic environment dynamics makes world models a good candidate for effective offline training. However, the effects of online vs. offline data on world models and thus on the resulting task performance have not been thoroughly studied in the literature. In this work, we investigate both paradigms in model-based settings, conducting experiments on 31 different environments. First, we showcase that online agents outperform their offline counterparts. We identify a key challenge behind performance degradation of offline agents: encountering Out-Of-Distribution states at test time. This issue arises because, without the self-correction mechanism in online agents, offline datasets with limited state space coverage induce a mismatch between the agent's imagination and real rollouts, compromising policy training. We demonstrate that this issue can be mitigated by allowing for additional online interactions in a fixed or adaptive schedule, restoring the performance of online training with limited interaction data. We also showcase that incorporating exploration data helps mitigate the performance degradation of offline agents. Based on our insights, we recommend adding exploration data when collecting large datasets, as current efforts predominantly focus on expert data alone.

We present a Reinforcement Learning (RL)-based locomotion system for Cosmo, a custom-built humanoid robot designed for entertainment applications. Unlike traditional humanoids, entertainment robots present unique challenges due to aesthetic-driven design choices. Cosmo embodies these with a disproportionately large head (16% of total mass), limited sensing, and protective shells that considerably restrict movement. To address these challenges, we apply Adversarial Motion Priors (AMP) to enable the robot to learn natural-looking movements while maintaining physical stability. We develop tailored domain randomization techniques and specialized reward structures to ensure safe sim-to-real, protecting valuable hardware components during deployment. Our experiments demonstrate that AMP generates stable standing and walking behaviors despite Cosmo's extreme mass distribution and movement constraints. These results establish a promising direction for robots that balance aesthetic appeal with functional performance, suggesting that learning-based methods can effectively adapt to aesthetic-driven design constraints.

Early diffusion models were formulated as continuous-time Markov chains over continuous spaces with Gaussian transition kernels (Sohl-Dickstein et al., 2015; Ho et al., 2020), and were later connected to continuous-time formulations via stochastic differential equations, offering a unifying perspective on score-based generative modeling (Song et al., 2020). In parallel, discrete diffusion has been developed from the viewpoint of Markov chains over discrete space (Hoogeboom et al., 2021). Notably, Austin et al. (2021) introduced D3PM with several families of discrete transition kernels, and Lou et al. (2023) proposed SEDD, which adopts score-based training objectives. A complementary line of work studies discrete flows (Campbell et al., 2024; Gat et al., 2024), aiming to understand continuous-time Markov chains (CTMCs) that interpolate between data and base distributions; this perspective aligns with ours. Subsequent extensions consider token-wise paths and path-wise structure within such flows (Shaul et al., 2024).

Multi-agent path finding (MAPF) is the problem of planning conflict-free paths from the designated start locations to goal positions for multiple agents. It underlies a variety of real-world tasks, including multi-robot coordination, robot-assisted logistics, and social navigation. Recent decentralized learnable solvers have shown great promise for large-scale MAPF, especially when leveraging foundation models and large datasets. However, these agents are reactive policy models and exhibit limited modeling of environmental temporal dynamics and inter-agent dependencies, resulting in performance degradation in complex, long-term planning scenarios. To address these limitations, we propose MAPF-World, an autoregressive action world model for MAPF that unifies situation understanding and action generation, guiding decisions beyond immediate local observations. It improves situational awareness by explicitly modeling environmental dynamics, including spatial features and temporal dependencies, through future state and actions prediction. By incorporating these predicted futures, MAPF-World enables more informed, coordinated, and far-sighted decision-making, especially in complex multi-agent settings. Furthermore, we augment MAPF benchmarks by introducing an automatic map generator grounded in real-world scenarios, capturing practical map layouts for training and evaluating MAPF solvers. Extensive experiments demonstrate that MAPF-World outperforms state-of-the-art learnable solvers, showcasing superior zero-shot generalization to out-of-distribution cases. Notably, MAPF-World is trained with a 96.5% smaller model size and 92% reduced data.

A universal controller for any robot morphology would greatly improve computational and data efficiency. By utilizing contextual information about the properties of individual robots and exploiting their modular structure in the architecture of deep reinforcement learning agents, steps have been made towards multi-robot control. Generalization to new, unseen robots, however, remains a challenge. In this paper we hypothesize that the relevant contextual information is partially observable, but that it can be inferred through interactions for better generalization to contexts that are not seen during training. To this extent, we implement a modular recurrent architecture and evaluate its generalization performance on a large set of MuJoCo robots. The results show a substantial improved performance on robots with unseen dynamics, kinematics, and topologies, in four different environments.

The iterative and stochastic nature of diffusion models enables test-time scaling, whereby spending additional compute during denoising generates higher-fidelity samples. Increasing the number of denoising steps is the primary scaling axis, but this yields quickly diminishing returns. Instead optimizing the noise trajectory--the sequence of injected noise vectors--is promising, as the specific noise realizations critically affect sample quality; but this is challenging due to a high-dimensional search space, complex noise-outcome interactions, and costly trajectory evaluations. We address this by first casting diffusion as a Markov Decision Process (MDP) with a terminal reward, showing tree-search methods such as Monte Carlo tree search (MCTS) to be meaningful but impractical. To balance performance and efficiency, we then resort to a relaxation of MDP, where we view denoising as a sequence of independent contextual bandits. This allows us to introduce an $ε$-greedy search algorithm that globally explores at extreme timesteps and locally exploits during the intermediate steps where de-mixing occurs. Experiments on EDM and Stable Diffusion reveal state-of-the-art scores for class-conditioned/text-to-image generation, exceeding baselines by up to $164\%$ and matching/exceeding MCTS performance. To our knowledge, this is the first practical method for test-time noise trajectory optimization of arbitrary (non-differentiable) rewards.