Wenbo Wu

Yue Wang, Yanran Xu, Wenbo Wu et al.

As robot control shifts toward large-scale reinforcement learning with in-loop dynamics computation, the community's reliance on CPU-bound libraries such as Pinocchio creates a throughput bottleneck in GPU-based training pipelines. We present BARD (Batched Articulated Rigid-body Dynamics), a self-contained PyTorch implementation of Featherstone's rigid-body dynamics algorithms, optimized for batched GPU evaluation and automatic differentiation. Three design choices make this efficient: a tiered lazy-evaluation cache that avoids redundant tree traversals, matmul-free joint transforms via pre-computed Rodrigues constants, and level-parallel propagation that reduces sequential operations to tree-depth batched steps. On five robot models (7-23 DOFs), BARD matches Pinocchio numerically while reaching up to 64x higher throughput for Forward Kinematics and 63x for Jacobians at batch size 4096 on an NVIDIA H200. We validate differentiability through gradient-based system identification on a 7-DOF manipulator, recovering link masses to 1.24% mean error under 5% torque noise, and integrate BARD into an Isaac Lab AMP training pipeline for an 11-DOF spined quadruped with 4096 parallel environments, where it is 8.5x faster than Pinocchio and 2.0x faster than ADAM for in-loop dynamics. BARD is open-sourced at: https://github.com/YueWang996/bard-pytorch-dynamics.

Shushu Yi, Wenbo Wu, Guoci Chen et al.

GPU has become the leading computing device for a wide range of data-intensive applications, which tightly collaborates with remote all-flash array (AFA) to accommodate ever-expanding datasets, facilitate multi-client data sharing, and guarantee fault tolerance. Although GPU is the center of computation, all I/O processes in existing GPU-AFA systems are still CPU-centric. CPU orchestrates remote I/O requests and executes a centralized AFA engine to take charge of AFA-level functionalities (e.g., access control and metadata persistence). This design disparity suffers from substantial CPU-GPU interaction overhead and I/O traffic amplification, compromising end-to-end I/O performance. In this work, we present \emph{GNStor}, a GPU-native AFA system that enables GPU to directly access remote AFA without CPU intervention in the I/O path, thereby fully exploiting the performance of AFA. Specifically, GNStor first proposes a GPU-centric NVMe over RDMA (NoR) software stack (named \emph{GNoR}), paving a fast path for GPUs to directly initiate NoR I/O requests to SSDs within remote AFA. GNoR employs an atomic-operation-based I/O orchestration design and follows the single-instruction-multiple-thread (SIMT) execution model of GPU, fully exploiting the massive parallelism of GPU architectures. To facilitate essential AFA functionalities in a CPU-bypass I/O path, GNStor further designs \emph{deEngine}, a decentralized AFA engine that seamlessly decomposes and integrates AFA-level tasks into each SSD firmware, thereby achieving efficient AFA access at low cost. Evaluation results show that GNStor achieves 3.2$\times$ higher I/O throughput and reduces application execution time by 31.1\%, compared to state-of-the-art AFA systems.

Yangjie Xu, Lujun Li, Lama Sleem et al.

Agent Skill framework, now widely and officially supported by major players such as GitHub Copilot, LangChain, and OpenAI, performs especially well with proprietary models by improving context engineering, reducing hallucinations, and boosting task accuracy. Based on these observations, an investigation is conducted to determine whether the Agent Skill paradigm provides similar benefits to small language models (SLMs). This question matters in industrial scenarios where continuous reliance on public APIs is infeasible due to data-security and budget constraints requirements, and where SLMs often show limited generalization in highly customized scenarios. This work introduces a formal mathematical definition of the Agent Skill process, followed by a systematic evaluation of language models of varying sizes across multiple use cases. The evaluation encompasses two open-source tasks and a real-world insurance claims data set. The results show that tiny models struggle with reliable skill selection, while moderately sized SLMs (approximately 12B - 30B) parameters) benefit substantially from the Agent Skill approach. Moreover, code-specialized variants at around 80B parameters achieve performance comparable to closed-source baselines while improving GPU efficiency. Collectively, these findings provide a comprehensive and nuanced characterization of the capabilities and constraints of the framework, while providing actionable insights for the effective deployment of Agent Skills in SLM-centered environments.

Rahul Ladhania, Jann Spiess, Lyle Ungar et al.

We consider learning personalized assignments to one of many treatment arms from a randomized controlled trial. Standard methods that estimate heterogeneous treatment effects separately for each arm may perform poorly in this case due to excess variance. We instead propose methods that pool information across treatment arms: First, we consider a regularized forest-based assignment algorithm based on greedy recursive partitioning that shrinks effect estimates across arms. Second, we augment our algorithm by a clustering scheme that combines treatment arms with consistently similar outcomes. In a simulation study, we compare the performance of these approaches to predicting arm-wise outcomes separately, and document gains of directly optimizing the treatment assignment with regularization and clustering. In a theoretical model, we illustrate how a high number of treatment arms makes finding the best arm hard, while we can achieve sizable utility gains from personalization by regularized optimization.

Wenbo Wu, George Konstantinidis

Trust and Reputation Management Systems (TRMSs) are critical for the modern web, yet their reliance on subjective user ratings or narrow Quality of Service (QoS) metrics lacks objective grounding. Concurrently, while regulatory frameworks like GDPR and HIPAA provide objective behavioral standards, automated compliance auditing has been limited to coarse, binary (pass/fail) outcomes. This paper bridges this research gap by operationalizing regulatory compliance as a quantitative and dynamic trust metric through our novel automated compliance engine (ACE). ACE first formalizes legal and organizational policies into a verifiable, obligation-centric logic. It then continuously audits system event logs against this logic to detect violations. The core of our contribution is a quantitative model that assesses the severity of each violation along multiple dimensions, including its Volume, Duration, Breadth, and Criticality, to compute a fine-grained, evolving compliance score. We evaluate ACE on a synthetic hospital dataset, demonstrating its ability to accurately detect a range of complex HIPAA and GDPR violations and produce a nuanced score that is significantly more expressive than traditional binary approaches. This work enables the development of more transparent, accountable, and resilient TRMSs on the Web.

Xianzhe Meng, Qiangsheng Zeng, Ling Luo et al.

Training stability is typically regarded as a prerequisite for reliable optimization in large language models. In this work, we analyze how stabilizing training dynamics affects the induced generation distribution. We show that under standard maximum likelihood training, stable parameter trajectories lead stationary solutions to approximately minimize the forward KL divergence to the empirical distribution, while implicitly reducing generative entropy. As a consequence, the learned model can concentrate probability mass on a limited subset of empirical modes, exhibiting systematic degeneration despite smooth loss convergence. We empirically validate this effect using a controlled feedback-based training framework that stabilizes internal generation statistics, observing consistent low-entropy outputs and repetitive behavior across architectures and random seeds. It indicates that optimization stability and generative expressivity are not inherently aligned, and that stability alone is an insufficient indicator of generative quality.

Wenbo Wu, Qingyi Si, Xiurui Pan et al.

While Key-Value (KV) cache succeeds in reducing redundant computations in auto-regressive models, it introduces significant memory overhead, limiting its practical deployment in long-sequence scenarios. Existing KV retrieval methods mitigate this by dynamically retaining only a subset of KV entries on the GPU. However, they still suffer from notable efficiency and accuracy bottlenecks due to per-token retrieval and coarse-grained page-level KV management, especially in long-output reasoning scenarios. With the emergence of large reasoning models, efficiently handling such scenarios has become increasingly important. To address this issue, we present two key observations: (1) critical KVs exhibit strong temporal locality during decoding, and (2) these KVs exhibit distinct distribution patterns across the input prompt and generated output. Building on these observations, we propose LouisKV, an efficient KV cache retrieval framework designed for various long-sequence scenarios. Specifically, LouisKV introduces a semantic-aware retrieval strategy leveraging temporal locality to trigger retrieval only at semantic boundaries, drastically reducing computation and data transfer overhead. LouisKV also designs a decoupled, fine-grained management scheme that tailors differentiated strategies for input and output sequences to create retrieval units that better match the model's attention patterns, enabling precise identification of critical KVs. Furthermore, to boost efficiency, LouisKV incorporates several kernel-level optimizations, including custom Triton and CUDA kernels to accelerate the KV clustering and retrieval. Evaluations show that LouisKV achieves up to 4.7$\times$ speedup over state-of-the-art KV retrieval methods while maintaining near-lossless accuracy across diverse long-sequence tasks, including long-input short-output, short-input long-output, and long-input long-output scenarios.