AI-based peer review systems tend to produce shallow and overpraising suggestions compared to human feedback. Here, we evaluate how well a reasoning LLM trained with multi-objective reinforcement learning (REMOR) can overcome these limitations. We start by designing a multi-aspect reward function that aligns with human evaluation of reviews. The aspects are related to the review itself (e.g., criticisms, novelty) and the relationship between the review and the manuscript (i.e., relevance). First, we perform supervised fine-tuning of DeepSeek-R1-Distill-Qwen-7B using LoRA on PeerRT, a new dataset of high-quality top AI conference reviews enriched with reasoning traces. We then apply Group Relative Policy Optimization (GRPO) to train two models: REMOR-H (with the human-aligned reward) and REMOR-U (with a uniform reward). Interestingly, the human-aligned reward penalizes aspects typically associated with strong reviews, leading REMOR-U to produce qualitatively more substantive feedback. Our results show that REMOR-U and REMOR-H achieve more than twice the average rewards of human reviews, non-reasoning state-of-the-art agentic multi-modal AI review systems, and general commercial LLM baselines. We found that while the best AI and human reviews are comparable in quality, REMOR avoids the long tail of low-quality human reviews. We discuss how reasoning is key to achieving these improvements and release the Human-aligned Peer Review Reward (HPRR) function, the Peer Review Reasoning-enriched Traces (PeerRT) dataset, and the REMOR models, which we believe can help spur progress in the area.

Unmanned Aerial Vehicles (UAVs) depend on onboard sensors for perception, navigation, and control. However, these sensors are susceptible to physical attacks, such as GPS spoofing, that can corrupt state estimates and lead to unsafe behavior. While reinforcement learning (RL) offers adaptive control capabilities, existing safe RL methods are ineffective against such attacks. We present ARMOR (Adaptive Robust Manipulation-Optimized State Representations), an attack-resilient, model-free RL controller that enables robust UAV operation under adversarial sensor manipulation. Instead of relying on raw sensor observations, ARMOR learns a robust latent representation of the UAV's physical state via a two-stage training framework. In the first stage, a teacher encoder, trained with privileged attack information, generates attack-aware latent states for RL policy training. In the second stage, a student encoder is trained via supervised learning to approximate the teacher's latent states using only historical sensor data, enabling real-world deployment without privileged information. Our experiments show that ARMOR outperforms conventional methods, ensuring UAV safety. Additionally, ARMOR improves generalization to unseen attacks and reduces training cost by eliminating the need for iterative adversarial training.

In the realm of online advertising, advertisers partake in ad auctions to obtain advertising slots, frequently taking advantage of auto-bidding tools provided by demand-side platforms. To improve the automation of these bidding systems, we adopt generative models, namely the Decision Transformer (DT), to tackle the difficulties inherent in automated bidding. Applying the Decision Transformer to the auto-bidding task enables a unified approach to sequential modeling, which efficiently overcomes short-sightedness by capturing long-term dependencies between past bidding actions and user behavior. Nevertheless, conventional DT has certain drawbacks: (1) DT necessitates a preset return-to-go (RTG) value before generating actions, which is not inherently produced; (2) The policy learned by DT is restricted by its training data, which is consists of mixed-quality trajectories. To address these challenges, we introduce the R* Decision Transformer (R* DT), developed in a three-step process: (1) R DT: Similar to traditional DT, R DT stores actions based on state and RTG value, as well as memorizing the RTG for a given state using the training set; (2) R^ DT: We forecast the highest value (within the training set) of RTG for a given state, deriving a suboptimal policy based on the current state and the forecasted supreme RTG value; (3) R* DT: Based on R^ DT, we generate trajectories and select those with high rewards (using a simulator) to augment our training dataset. This data enhancement has been shown to improve the RTG of trajectories in the training data and gradually leads the suboptimal policy towards optimality. Comprehensive tests on a publicly available bidding dataset validate the R* DT's efficacy and highlight its superiority when dealing with mixed-quality trajectories.

Laser powder bed fusion (L-PBF) is a widely recognized additive manufacturing technology for producing intricate metal components with exceptional accuracy. A key challenge in L-PBF is the formation of complex microstructures affecting product quality. We propose a physics-guided, machine-learning approach to optimize scan paths for desired microstructure outcomes, such as equiaxed grains. We utilized a phase-field method (PFM) to model crystalline grain structure evolution. To reduce computational costs, we trained a surrogate machine learning model, a 3D U-Net convolutional neural network, using single-track phase-field simulations with various laser powers to predict crystalline grain orientations based on initial microstructure and thermal history. We investigated three scanning strategies across various hatch spacings within a square domain, achieving a two-orders-of-magnitude speedup using the surrogate model. To reduce trial and error in designing laser scan toolpaths, we used deep reinforcement learning (DRL) to generate optimized scan paths for target microstructure. Results from three cases demonstrate the DRL approach's effectiveness. We integrated the surrogate 3D U-Net model into our DRL environment to accelerate the reinforcement learning training process. The reward function minimizes both aspect ratio and grain volume of the predicted microstructure from the agent's scan path. The reinforcement learning algorithm was benchmarked against conventional zigzag approach for smaller and larger domains, showing machine learning methods' potential to enhance microstructure control and computational efficiency in L-PBF optimization.

Deep reinforcement learning has been extensively studied in decision-making processes and has demonstrated superior performance over conventional approaches in various fields, including radar resource management (RRM). However, a notable limitation of neural networks is their ``black box" nature and recent research work has increasingly focused on explainable AI (XAI) techniques to describe the rationale behind neural network decisions. One promising XAI method is local interpretable model-agnostic explanations (LIME). However, the sampling process in LIME ignores the correlations between features. In this paper, we propose a modified LIME approach that integrates deep learning (DL) into the sampling process, which we refer to as DL-LIME. We employ DL-LIME within deep reinforcement learning for radar resource management. Numerical results show that DL-LIME outperforms conventional LIME in terms of both fidelity and task performance, demonstrating superior performance with both metrics. DL-LIME also provides insights on which factors are more important in decision making for radar resource management.

Unsupervised skill discovery aims to learn diverse and distinguishable behaviors in open-ended reinforcement learning. For existing methods, they focus on improving diversity through pure exploration, mutual information optimization, and learning temporal representation. Despite that they perform well on exploration, they remain limited in terms of efficiency, especially for the high-dimensional situations. In this work, we frame skill discovery as a min-max game of skill generation and policy learning, proposing a regret-aware method on top of temporal representation learning that expands the discovered skill space along the direction of upgradable policy strength. The key insight behind the proposed method is that the skill discovery is adversarial to the policy learning, i.e., skills with weak strength should be further explored while less exploration for the skills with converged strength. As an implementation, we score the degree of strength convergence with regret, and guide the skill discovery with a learnable skill generator. To avoid degeneration, skill generation comes from an up-gradable population of skill generators. We conduct experiments on environments with varying complexities and dimension sizes. Empirical results show that our method outperforms baselines in both efficiency and diversity. Moreover, our method achieves a 15% zero shot improvement in high-dimensional environments, compared to existing methods.

We investigate the effectiveness of reinforcement learning methods for finetuning large language models when transitioning from offline to semi-online to fully online regimes for both verifiable and non-verifiable tasks. Our experiments cover training on verifiable math as well as non-verifiable instruction following with a set of benchmark evaluations for both. Across these settings, we extensively compare online and semi-online Direct Preference Optimization and Group Reward Policy Optimization objectives, and surprisingly find similar performance and convergence between these variants, which all strongly outperform offline methods. We provide a detailed analysis of the training dynamics and hyperparameter selection strategies to achieve optimal results. Finally, we show that multi-tasking with verifiable and non-verifiable rewards jointly yields improved performance across both task types.

With a simple auxiliary loss, that equalizes these distributions via MSE, rQdia boosts DrQ and SAC on 9/ 12 and 10 /12 tasks respectively in the MuJoCo Continuous Control Suite from pixels, and Data-Efficient Rainbow on 18/ 26 Atari Arcade environments. Gains are measured in both sample efficiency and longer-term training. Human perception is invariant to and remarkably robust against many perturbations, like discoloration, obfuscation, and low exposure. On the other hand, artificial neural networks do not intrinsically carry these invariance properties, though some invariances may be induced architecturally through inductive biases like convolution, kernel rotation, and dilation. In deep reinforcement learning (RL) from pixels, an agent is tasked to learn from raw pixels and must therefore learn to visually interpret a scene. Thus, recent approaches in deep RL have turned to the self-supervision and data augmentation techniques found in computer vision.

--Reinforcement learning (RL) policies deployed in safety-critical systems, such as unmanned aerial vehicle (UA V) navigation in dynamic airspace, are vulnerable to out-of-distribution (OOD) adversarial attacks in the observation space. These attacks induce distributional shifts that significantly degrade value estimation, leading to unsafe or suboptimal decision-making rendering the existing policy fragile. T o address this vulnerability, we propose an antifragile RL framework designed to adapt against curriculum of incremental adversarial perturbations. The framework introduces a simulated attacker which incrementally increases the strength of observation-space perturbations which enables the RL agent to adapt and generalize across a wider range of OOD observations and anticipate previously unseen attacks. We begin with a theoretical characterization of fragility, formally defining catastrophic forgetting as a monotonic divergence in value function distributions with increasing perturbation strength. Building on this, we define antifragility as the boundedness of such value shifts and derive adaptation conditions under which forgetting is stabilized. Our method enforces these bounds through iterative expert-guided critic alignment using Wasserstein distance minimization across incrementally perturbed observations. We empirically evaluate the approach in a UA V deconfliction scenario involving dynamic 3D obstacles. Results show that the antifragile policy consistently outperforms standard and robust RL baselines when subjected to both projected gradient descent (PGD) and GPS spoofing attacks, achieving up to 15% higher cumulative reward and over 30% fewer conflict events. These findings demonstrate the practical and theoretical viability of antifragile reinforcement learning for secure and resilient decision-making in environments with evolving threat scenarios. Fragility, robustness, and antifragility represent a continuum of system responses to stress or external perturbations [1, 2].

Trajectory prediction for surrounding agents is a challenging task in autonomous driving due to its inherent uncertainty and underlying multimodality. Unlike prevailing data-driven methods that primarily rely on supervised learning, in this paper, we introduce a novel Graph-oriented Inverse Reinforcement Learning (GoIRL) framework, which is an IRL-based predictor equipped with vectorized context representations. We develop a feature adaptor to effectively aggregate lane-graph features into grid space, enabling seamless integration with the maximum entropy IRL paradigm to infer the reward distribution and obtain the policy that can be sampled to induce multiple plausible plans. Furthermore, conditioned on the sampled plans, we implement a hierarchical parameterized trajectory generator with a refinement module to enhance prediction accuracy and a probability fusion strategy to boost prediction confidence. Extensive experimental results showcase our approach not only achieves state-of-the-art performance on the large-scale Argoverse & nuScenes motion forecasting benchmarks but also exhibits superior generalization abilities compared to existing supervised models.