-- We present Kalman-filter Assisted Reinforcement Learner (KARL) for dynamic object tracking and grasping over eye-on-hand (EoH) systems, significantly expanding such systems' capabilities in challenging, realistic environments. In comparison to the previous state-of-the-art, KARL (1) incorporates a novel six-stage RL curriculum that doubles the system's motion range, thereby greatly enhancing the system's grasping performance, (2) integrates a robust Kalman filter layer between the perception and reinforcement learning (RL) control modules, enabling the system to maintain an uncertain but continuous 6D pose estimate even when the target object temporarily exits the camera's field-of-view or undergoes rapid, unpredictable motion, and (3) introduces mechanisms to allow retries to gracefully recover from unavoidable policy execution failures. Extensive evaluations conducted in both simulation and real-world experiments qualitatively and quantitatively corroborate KARL's advantage over earlier systems, achieving higher grasp success rates and faster robot execution speed. Source code and supplementary materials for KARL will be made available at: https://github.com/arc-l/karl . Humans, and animals in general, interact with the physical world through observing and handling everyday objects [1], which makes object tracking and manipulation arguably the most fundamental skill for physical intelligence. In robotics, autonomous grasping in stationary settings has been extensively studied [2], [3], typically using decoupled vision and manipulation sub-systems where the camera does not move with the manipulator. While effective for static tasks, this approach struggles in dynamic scenarios where objects move or become occluded. Real-world interactions, such as handovers, require continuous tracking and adaptive grasping, highlighting the need for more integrated solutions.

In this paper, we first propose a novel algorithm for model fusion that leverages Wasserstein barycenters in training a global Deep Neural Network (DNN) in a distributed architecture. To this end, we divide the dataset into equal parts that are fed to "agents" who have identical deep neural networks and train only over the dataset fed to them (known as the local dataset). After some training iterations, we perform an aggregation step where we combine the weight parameters of all neural networks using Wasserstein barycenters. These steps form the proposed algorithm referred to as FedWB. Moreover, we leverage the processes created in the first part of the paper to develop an algorithm to tackle Heterogeneous Federated Reinforcement Learning (HFRL). Our test experiment is the CartPole toy problem, where we vary the lengths of the poles to create heterogeneous environments. We train a deep Q-Network (DQN) in each environment to learn to control each cart, while occasionally performing a global aggregation step to generalize the local models; the end outcome is a global DQN that functions across all environments.

Compiler auto-tuning optimizes pass sequences to improve performance metrics such as Intermediate Representation (IR) instruction count. Although recent advances leveraging Large Language Models (LLMs) have shown promise in automating compiler tuning, two significant challenges still remain: the absence of high-quality reasoning datasets for agents training, and limited effective interactions with the compilation environment. In this work, we introduce Compiler-R1, the first reinforcement learning (RL)-driven framework specifically augmenting LLM capabilities for compiler auto-tuning. Compiler-R1 features a curated, high-quality reasoning dataset and a novel two-stage end-to-end RL training pipeline, enabling efficient environment exploration and learning through an outcome-based reward. Extensive experiments across seven datasets demonstrate Compiler-R1 achieving an average 8.46% IR instruction count reduction compared to opt -Oz, showcasing the strong potential of RL-trained LLMs for compiler optimization. Our code and datasets are publicly available at https://github.com/Panhaolin2001/Compiler-R1.

Reinforcement learning (RL) agents are capable of making complex decisions in dynamic environments, yet their behavior often remains opaque. When an agent executes a sequence of actions--such as administering insulin to a diabetic patient or controlling a spacecraft's landing--it is rarely clear how outcomes might have changed under alternative choices. This challenge becomes particularly pronounced in settings involving continuous action spaces, where decisions are not confined to discrete options but span a spectrum of real-valued magnitudes. The framework introduced in recent work aims to generate counterfactual explanations in such settings, offering a structured approach to explore "what if" scenarios. The value of counterfactual reasoning in RL becomes apparent in scenarios with high-stakes, temporally extended consequences.

Generating high-quality and photorealistic 3D assets remains a longstanding challenge in 3D vision and computer graphics. Although state-of-the-art generative models, such as diffusion models, have made significant progress in 3D generation, they often fall short of human-designed content due to limited ability to follow instructions, align with human preferences, or produce realistic textures, geometries, and physical attributes. In this paper, we introduce Nabla-R2D3, a highly effective and sample-efficient reinforcement learning alignment framework for 3D-native diffusion models using 2D rewards. Built upon the recently proposed Nabla-GFlowNet method, which matches the score function to reward gradients in a principled manner for reward finetuning, our Nabla-R2D3 enables effective adaptation of 3D diffusion models using only 2D reward signals. Extensive experiments show that, unlike vanilla finetuning baselines which either struggle to converge or suffer from reward hacking, Nabla-R2D3 consistently achieves higher rewards and reduced prior forgetting within a few finetuning steps.

Inspired by the impressive reasoning capabilities demonstrated by reinforcement learning approaches like DeepSeek-R1, recent emerging research has begun exploring the use of reinforcement learning (RL) to enhance vision-language models (VLMs) for multimodal reasoning tasks. However, most existing multimodal reinforcement learning approaches remain limited to spatial reasoning within single-image contexts, yet still struggle to generalize to more complex and real-world scenarios involving multi-image positional reasoning, where understanding the relationships across images is crucial. To address this challenge, we propose a general reinforcement learning approach PeRL tailored for interleaved multimodal tasks, and a multi-stage strategy designed to enhance the exploration-exploitation trade-off, thereby improving learning efficiency and task performance. Specifically, we introduce permutation of image sequences to simulate varied positional relationships to explore more spatial and positional diversity. Furthermore, we design a rollout filtering mechanism for resampling to focus on trajectories that contribute most to learning optimal behaviors to exploit learned policies effectively. We evaluate our model on 5 widely-used multi-image benchmarks and 3 single-image benchmarks. Our experiments confirm that PeRL trained model consistently surpasses R1-related and interleaved VLM baselines by a large margin, achieving state-of-the-art performance on multi-image benchmarks, while preserving comparable performance on single-image tasks.

In reinforcement learning (RL), agents continually interact with the environment and use the feedback to refine their behavior. To guide policy optimization, reward models are introduced as proxies of the desired objectives, such that when the agent maximizes the accumulated reward, it also fulfills the task designer's intentions. Recently, significant attention from both academic and industrial researchers has focused on developing reward models that not only align closely with the true objectives but also facilitate policy optimization. In this survey, we provide a comprehensive review of reward modeling techniques within the deep RL literature. We begin by outlining the background and preliminaries in reward modeling. Next, we present an overview of recent reward modeling approaches, categorizing them based on the source, the mechanism, and the learning paradigm. Building on this understanding, we discuss various applications of these reward modeling techniques and review methods for evaluating reward models. Finally, we conclude by highlighting promising research directions in reward modeling. Altogether, this survey includes both established and emerging methods, filling the vacancy of a systematic review of reward models in current literature.

-- We propose a hierarchical reinforcement learning (HRL) framework for efficient Navigation Among Movable Obstacles ( NAMO) using a mobile manipulator . Our approach combines interaction-based obstacle property estimation with structured pushing strategies, facilitating the dynamic manipulation of unforeseen obstacles while adhering to a pre-planned global path. The high-level policy generates pushing commands that consider environmental constraints and path-tracking objectives, while the low-level policy precisely and stably executes these commands through coordinated whole-body movements. Comprehensive simulation-based experiments demonstrate improvements in performing NAMO tasks, including higher success rates, shortened traversed path length, and reduced goal-reaching times, compared to baselines. Additionally, ablation studies assess the efficacy of each component, while a qualitative analysis further validates the accuracy and reliability of the real-time obstacle property estimation. Robust robot navigation in complex environments is crucial for applications ranging from delivery [1] to warehouse automation [2].

Scaling deep reinforcement learning networks is challenging and often results in degraded performance, yet the root causes of this failure mode remain poorly understood. Several recent works have proposed mechanisms to address this, but they are often complex and fail to highlight the causes underlying this difficulty. In this work, we conduct a series of empirical analyses which suggest that the combination of non-stationarity with gradient pathologies, due to suboptimal architectural choices, underlie the challenges of scale. We propose a series of direct interventions that stabilize gradient flow, enabling robust performance across a range of network depths and widths. Our interventions are simple to implement and compatible with well-established algorithms, and result in an effective mechanism that enables strong performance even at large scales. We validate our findings on a variety of agents and suites of environments.

Reinforcement learning from human feedback (RLHF) has become a key factor in aligning model behavior with users' goals. However, while humans integrate multiple strategies when making decisions, current RLHF approaches often simplify this process by modeling human reasoning through isolated tasks such as classification or regression. In this paper, we propose a novel reinforcement learning (RL) method that mimics human decision-making by jointly considering multiple tasks. Specifically, we leverage human ratings in reward-free environments to infer a reward function, introducing learnable weights that balance the contributions of both classification and regression models. This design captures the inherent uncertainty in human decision-making and allows the model to adaptively emphasize different strategies. We conduct several experiments using synthetic human ratings to validate the effectiveness of the proposed approach. Results show that our method consistently outperforms existing rating-based RL methods, and in some cases, even surpasses traditional RL approaches.