arXiv subjects

Wenyi Wang

Publications and source records attributed to Wenyi Wang.

At least 19 recordsLinked to original sources

arXiv2026

Directed partial orders on the complex number field

We construct a class of positive cones that make $\C$ into a directed partially ordered ring. The positive cones are defined using integral closures of local rings, associated with a transcendence basis and a chosen real generator. A localization criterion also yields such orders on every transcendental extension of $\Q$. For a fixed coefficient field, we prove that two real generators define the same cone if and only if they differ by an affine map with positive real-algebraic slope and translation algebraic over that field. The nonzero positive elements are closed under inversion. None of these orders is a lattice order.

math.RA↗

arXiv2026

HyGra: Accelerating Network-State Simulation for LLM Training in DCNs via Adaptive Packet-Flow Granularity

In recent years, large language models (LLMs) have driven substantial intelligent transformation across diverse industries. Commercial LLM training is typically performed over data center networks (DCNs) comprising hundreds to thousands of GPUs, with multiple devices collocated per node. As network scale expands, inter-node communication becomes a primary bottleneck to training efficiency. Network-state simulators therefore play a crucial role by enabling cost-effective evaluation of network configurations and parallelization strategies through faithful emulation of DCN dynamics during LLM training. However, existing simulators are constrained by a efficiency-fidelity tradeoff, as packet-level simulators (PLSs) incur prohibitive runtime overhead, whereas flow-level simulators (FLSs) compromise essential modeling accuracy. In this paper, we develop \texttt{HyGra}, a hybrid-granularity network-state simulator that exploits intrinsic network dynamics in LLM training to adaptively switch simulation granularity. Specifically, \texttt{HyGra} employs packet-level simulation during non-steady phases with transient fluctuations and flow-level simulation during steady phases with periodic patterns, thereby accelerating execution while preserving high fidelity. Moreover, it requires no specialized hardware, supports single-machine deployment, and is compatible with existing simulators. Experiments based representative commercial LLM workloads, including ChatGPT, DeepSeek, and Qwen, show that \texttt{HyGra} achieves up to 15.4$\times$ speedup under single parallelization strategy and 7.8$\times$ under hybrid parallelization strategies while maintaining high accuracy.

cs.NI↗

arXiv2026

ExaServe: Large-Scale LLM Serving on Exascale HPC Systems

Cloud-native LLM serving frameworks have made deployment routine in data centers, yet deploying them on leadership-class supercomputers remains an engineering challenge requiring scheduler integration, MPI launch, accelerator selection, node-local weight staging, and platform-specific patches. We present ExaServe, a pip-installable framework that transforms a declarative YAML specification into a reproducible large-scale LLM serving deployment. Using ExaServe, we deploy LLM serving on ALCF Aurora from 1 to 256 nodes (3072 vLLM replicas). Non-streaming inference scales nearly linearly to 256 nodes, reaching 27.1k requests/s (3.8M tokens/s). Token streaming scales differently: a centralized proxy plateaus at about 4.7k requests/s despite the model servers remaining within the service-level objective. We also identify an O(N^2) Ray Serve control-plane bottleneck that increases cluster bring-up to roughly 30 minutes at 256 nodes. ExaServe provides a practical, reproducible deployment path while exposing key barriers to future exascale LLM serving.

cs.DC↗

arXiv2026

A functional joint model with baseline functional covariates: linking sitting accumulation patterns to physical function and mortality among older women

In large-scale epidemiological studies, it is often of interest to investigate joint relationships between longitudinal and time-to-event outcomes with exposures that are trajectories or functions. Our motivation study is the Objective Physical Activity and Cardiovascular Health (OPACH) Study, which collected accelerometry-measured physical activity in 6,489 older women. One of the scientific aims is to understand sedentary behavior accumulation patterns and its association with physical function (longitudinal) and mortality (time-to-event). We propose a novel approach that first converts raw accelerometry data into daily sitting bout accumulation profiles, which are treated as functional covariates, and then develop a functional joint model for longitudinal and time-to-event outcomes that incorporates a baseline functional covariate. The longitudinal process is modeled using functional data methods and linked to the survival process through functional principal component scores. Both sub-models include interpretable linear scalar-on-function regression coefficients to capture flexible dose-response associations between sitting accumulated across varying bout durations and health outcomes. Estimation is carried out via an efficient expectation-maximization (EM) algorithm with penalized spline approximations. Simulation studies demonstrate accurate parameter estimation and reliable model selection. Application to the OPACH data reveals flexible and interpretable dose-response relationships between sitting bout durations, physical function, and mortality.

stat.AP↗

arXiv2026

SciDiagramEdit: Learning to Edit Scientific Diagrams from Paper Revisions

Editing the figures in a research paper is a routine and time-consuming part of everyday research practice: authors relabel components, rearrange panels, and restyle visuals as they revise their manuscripts. Automating this editing workflow under a natural-language instruction, however, is challenging, because a scientific figure is a dense infographic in which heterogeneous visual elements such as schematics, plots, photos, captions, and arrows are composed under a tight visual grammar to advance a specific argument. To address this, we present SciDiagramEdit, a benchmark and skill-evolution framework that learns from natural paper revisions and operates on the figure's editable vector source, where users can inspect and co-edit individual primitives alongside the agent. Our benchmark mines before/after figure pairs from arXiv version histories, each grounded in the authors' own revision intent. To accommodate the diversity of editing instructions, we adopt agentic learning via skill evolution: an agentic proposer continually refines the agent's skill specification from execution traces over multiple epochs. The resulting skill progressively lifts edit accuracy on a held-out validation set, providing evidence that natural paper revisions are an effective training signal for instruction-driven figure editing.

cs.CL↗

arXiv2026

Self-Improvements in Modern Agentic Systems: A Survey

Self-improving autonomous agents are moving from research prototypes to deployed systems. The primary goal is controllable evolution, or adaptation, from experience with minimal or even no human input. This survey frames modern self-improving agents as adaptive systems that convert experience into accumulated capability gains. We offer a system-level framework that represents a modern agent as a configuration coupling a foundation model with an operational scaffold of prompts, memory, tools, and control logic. Within this framework, self-improvement is formalized as a self-induced update operator that obtains and commits updates to model parameters or scaffold components. We organize prior work by update target and by the signals that drive change, then review applications and discuss evaluation, before closing with open problems and future directions. For convenience, we track technical updates on https://github.com/selfimproving-agent/awesome-Self-Improving-Agents.

cs.AI↗

arXiv2026

Neural Computers

We propose a new frontier: Neural Computers (NCs) that unify computation, memory, and I/O of traditional computers in a learned runtime state. Our long-term goal is the Completely Neural Computer (CNC): the mature, general-purpose realization of this emerging machine form, with stable execution, explicit reprogramming, and durable capability reuse. As an initial step, we study whether elementary NC primitives can be learned solely from collected I/O traces, without instrumented program state. Concretely, we instantiate NCs as video models that roll out screen frames from instructions, pixels, and user actions (when available) in CLI and GUI settings. We show that NCs can acquire elementary interface primitives, especially I/O alignment and short-horizon control, while routine reuse, controlled updates, and symbolic stability remain challenging. We outline a roadmap toward CNCs, to establish a new computing paradigm beyond today's agents and conventional computers.

cs.LG↗

arXiv2026

Scalar-Induced Electromagnetic Radiation: Comparison with Axion-Like Particles and Implications for Modified Gravity

The scalar-tensor theory of gravity, a modified gravity theory, introduces a fundamental scalar field that can serve as dynamical dark energy, driving the late-time accelerated expansion of the Universe. In this work, we analyze electromagnetic (EM) radiations arising from scalar fields and compare these features with those induced by axion-like particles (ALPs). Scalar and ALP fields couple differently to the EM field due to their distinct parity properties, $ϕF_{μν} F^{μν}$ for scalar fields and $ϕF_{μν} \tilde{F}^{μν}$ for ALPs. Building on analytical methods developed for ALPs, this work presents a theoretical feasibility analysis that demonstrates how the scalar field could produce observable EM signatures from oscillating field configurations. We also show that resonance effects can amplify the EM radiation for the scalar field under specific conditions, and that the enhancement mechanisms depend on the coupling structure and the configuration of the background magnetic field. Resonance phenomena can accentuate the differences in signal strength and spectral features, potentially aiding future observations in distinguishing scalar fields from ALPs. Although our studies apply to general scalar fields, we embed them within the framework of scalar-tensor theory and discuss the mass and coupling parameter in the context of testing modified gravity. This work provides a theoretical framework for studying generic pure and pseudo-scalar fields on an equal footing and suggests new avenues for observational tests of modified gravity scenarios alongside ALP models.

hep-ph↗

arXiv2026

Planning to Explore: Curiosity-Driven Planning for LLM Test Generation

The use of LLMs for code generation has naturally extended to code testing and evaluation. As codebases grow in size and complexity, so does the need for automated test generation. Current approaches for LLM-based test generation rely on strategies that maximize immediate coverage gain, a greedy approach that plateaus on code where reaching deep branches requires setup steps that individually yield zero new coverage. Drawing on principles of Bayesian exploration, we treat the program's branch structure as an unknown environment, and an evolving coverage map as a proxy probabilistic posterior representing what the LLM has discovered so far. Our method, CovQValue, feeds the coverage map back to the LLM, generates diverse candidate plans in parallel, and selects the most informative plan by LLM-estimated Q-values, seeking actions that balance immediate branch discovery with future reachability. Our method outperforms greedy selection on TestGenEval Lite, achieving 51-77% higher branch coverage across three popular LLMs and winning on 77-84% of targets. In addition, we build a benchmark for iterative test generation, RepoExploreBench, where they achieve 40-74%. These results show the potential of curiosity-driven planning methods for LLM-based exploration, enabling more effective discovery of program behavior through sequential interaction

cs.SE↗

arXiv2026

Mindstorms in Natural Language-Based Societies of Mind

Both Minsky's "society of mind" and Schmidhuber's "learning to think" inspire diverse societies of large multimodal neural networks (NNs) that solve problems by interviewing each other in a "mindstorm." Recent implementations of NN-based societies of minds consist of large language models (LLMs) and other NN-based experts communicating through a natural language interface. In doing so, they overcome the limitations of single LLMs, improving multimodal zero-shot reasoning. In these natural language-based societies of mind (NLSOMs), new agents -- all communicating through the same universal symbolic language -- are easily added in a modular fashion. To demonstrate the power of NLSOMs, we assemble and experiment with several of them (having up to 129 members), leveraging mindstorms in them to solve some practical AI tasks: visual question answering, image captioning, text-to-image synthesis, 3D generation, egocentric retrieval, embodied AI, and general language-based task solving. We view this as a starting point towards much larger NLSOMs with billions of agents-some of which may be humans. And with this emergence of great societies of heterogeneous minds, many new research questions have suddenly become paramount to the future of artificial intelligence. What should be the social structure of an NLSOM? What would be the (dis)advantages of having a monarchical rather than a democratic structure? How can principles of NN economies be used to maximize the total reward of a reinforcement learning NLSOM? In this work, we identify, discuss, and try to answer some of these questions.

cs.AI↗

arXiv2026

Learning Glioblastoma Tumor Heterogeneity Using Brain Inspired Topological Neural Networks

Accurate prognosis for Glioblastoma (GBM) using deep learning (DL) is hindered by extreme spatial and structural heterogeneity. Moreover, inconsistent MRI acquisition protocols across institutions hinder generalizability of models. Conventional transformer and DL pipelines often fail to capture the multi-scale morphological diversity such as fragmented necrotic cores, infiltrating margins, and disjoint enhancing components leading to scanner-specific artifacts and poor cross-site prognosis. We propose TopoGBM, a learning framework designed to capture heterogeneity-preserved, scanner-robust representations from multi-parametric 3D MRI. Central to our approach is a 3D convolutional autoencoder regularized by a topological regularization that preserves the complex, non-Euclidean invariants of the tumor's manifold within a compressed latent space. By enforcing these topological priors, TopoGBM explicitly models the high-variance structural signatures characteristic of aggressive GBM. Evaluated across heterogeneous cohorts (UPENN, UCSF, RHUH) and external validation on TCGA, TopoGBM achieves better performance (C-index 0.67 test, 0.58 validation), outperforming baselines that degrade under domain shift. Mechanistic interpretability analysis reveals that reconstruction residuals are highly localized to pathologically heterogeneous zones, with tumor-restricted and healthy tissue error significantly low (Test: 0.03, Validation: 0.09). Furthermore, occlusion-based attribution localizes approximately 50% of the prognostic signal to the tumor and the diverse peritumoral microenvironment advocating clinical reliability of the unsupervised learning method. Our findings demonstrate that incorporating topological priors enables the learning of morphology-faithful embeddings that capture tumor heterogeneity while maintaining cross-institutional robustness.

cs.LG↗

arXiv2026

Stimulated radiation from superradiant scalar cloud in scalar-tensor theory

Scalar-tensor theories predict fundamental scalar fields of considerable interest in astrophysics and cosmology. We investigate the superradiant instability of scalar clouds around Kerr black holes, showing that stimulated decay generates detectable electromagnetic signals. The growth of the superradiant scalar cloud differs from that of other bosonic fields and depends sensitively on the matter distribution surrounding the black hole, which originates from the scalar-matter coupling realized by the chameleon mechanism in modified gravity theories. In non-uniform matter distributions, stimulated emission from scalar clouds offers an observational signature that distinguishes fundamental scalars from other light bosonic fields.

gr-qc↗

arXiv2025

Random Regularity of the Vlasov-Poisson System with Random Initial Inputs in the Quasineutral Regime

The Vlasov-Poisson system is widely used in plasma physics and other related fields. In this paper, we study the Vlasov-Poisson system with initial uncertainty in the quasineutral regime. First, we prove the uniform convergence in the Wasserstein distance between the uncertain Vlasov-Poisson system in the quasineutral regime and its quasineutral limit system with random initial inputs. This is achieved by deriving an upper bound for the Wasserstein distance and rigorously estimating each component of this bound. Furthermore, by defining a new norm with respect to the quasineutral parameter and estimating the distribution function as well as the electric field in this norm using a variable substitution, we establish the random regularity of the solutions in the quasineutral regime. This work develops a novel framework for quantifying the propagation of the initial uncertainty of the Vlasov-Poisson system in the quasineutral regime, providing a theoretical basis for designing high-performance numerical algorithms.

math.AP↗

arXiv2025

Huxley-Gödel Machine: Human-Level Coding Agent Development by an Approximation of the Optimal Self-Improving Machine

Recent studies operationalize self-improvement through coding agents that edit their own codebases. They grow a tree of self-modifications through expansion strategies that favor higher software engineering benchmark performance, assuming that this implies more promising subsequent self-modifications. However, we identify a mismatch between the agent's self-improvement potential (metaproductivity) and its coding benchmark performance, namely the Metaproductivity-Performance Mismatch. Inspired by Huxley's concept of clade, we propose a metric ($\mathrm{CMP}$) that aggregates the benchmark performances of the descendants of an agent as an indicator of its potential for self-improvement. We show that, in our self-improving coding agent development setting, access to the true $\mathrm{CMP}$ is sufficient to simulate how the Gödel Machine would behave under certain assumptions. We introduce the Huxley-Gödel Machine (HGM), which, by estimating $\mathrm{CMP}$ and using it as guidance, searches the tree of self-modifications. On SWE-bench Verified and Polyglot, HGM outperforms prior self-improving coding agent development methods while using fewer allocated CPU hours. Last but not least, HGM demonstrates strong transfer to other coding datasets and large language models. The agent optimized by HGM on SWE-bench Verified with GPT-5-mini and evaluated on SWE-bench Lite with GPT-5 achieves human-level performance, matching the best officially checked results of human-engineered coding agents. Our code is publicly available at https://github.com/metauto-ai/HGM.

cs.AI↗

arXiv2025

Optimizing Fine-Grained Parallelism Through Dynamic Load Balancing on Multi-Socket Many-Core Systems

Achieving efficient task parallelism on many-core architectures is an important challenge. The widely used GNU OpenMP implementation of the popular OpenMP parallel programming model incurs high overhead for fine-grained, short-running tasks due to time spent on runtime synchronization. In this work, we introduce and analyze three key advances that collectively achieve significant performance gains. First, we introduce XQueue, a lock-less concurrent queue implementation to replace GNU's priority task queue and remove the global task lock. Second, we develop a scalable, efficient, and hybrid lock-free/lock-less distributed tree barrier to address the high hardware synchronization overhead from GNU's centralized barrier. Third, we develop two lock-less and NUMA-aware load balancing strategies. We evaluate our implementation using Barcelona OpenMP Task Suite (BOTS) benchmarks. We show that the use of XQueue and the distributed tree barrier can improve performance by up to 1522.8$\times$ compared to the original GNU OpenMP. We further show that lock-less load balancing can improve performance by up to 4$\times$ compared to GNU OpenMP using XQueue.

cs.DC↗

arXiv2025

FACTS: A Factored State-Space Framework For World Modelling

World modelling is essential for understanding and predicting the dynamics of complex systems by learning both spatial and temporal dependencies. However, current frameworks, such as Transformers and selective state-space models like Mambas, exhibit limitations in efficiently encoding spatial and temporal structures, particularly in scenarios requiring long-term high-dimensional sequence modelling. To address these issues, we propose a novel recurrent framework, the \textbf{FACT}ored \textbf{S}tate-space (\textbf{FACTS}) model, for spatial-temporal world modelling. The FACTS framework constructs a graph-structured memory with a routing mechanism that learns permutable memory representations, ensuring invariance to input permutations while adapting through selective state-space propagation. Furthermore, FACTS supports parallel computation of high-dimensional sequences. We empirically evaluate FACTS across diverse tasks, including multivariate time series forecasting, object-centric world modelling, and spatial-temporal graph prediction, demonstrating that it consistently outperforms or matches specialised state-of-the-art models, despite its general-purpose world modelling design.

cs.AI↗

arXiv2024

Solar system tests in covariant f(Q) gravity

We study the Solar System constraints on covariant $f(Q)$ gravity. The covariant $f(Q)$ theory is described by the metric and affine connection, where both the torsion and curvature vanish. Considering a model including a higher nonmetricity-scalar correction, $f(Q)= Q +αQ^{n} - 2Λ$, we derive static and spherically symmetric solutions, which represent the Schwarzschild-de Sitter solution with higher-order corrections, for two different ansatz of the affine connection. On the obtained spacetime solutions, we investigate the perihelion precession, light deflection, Shapiro delay, Cassini constraint, and gravitational redshift in the $f(Q)$ gravity. We place bounds on the parameter $α$ with $n=2, 3$ in our model of $f(Q)$ gravity, using various observational data in the Solar System.

gr-qc↗

arXiv2024

How to Correctly do Semantic Backpropagation on Language-based Agentic Systems

Language-based agentic systems have shown great promise in recent years, transitioning from solving small-scale research problems to being deployed in challenging real-world tasks. However, optimizing these systems often requires substantial manual labor. Recent studies have demonstrated that these systems can be represented as computational graphs, enabling automatic optimization. Despite these advancements, most current efforts in Graph-based Agentic System Optimization (GASO) fail to properly assign feedback to the system's components given feedback on the system's output. To address this challenge, we formalize the concept of semantic backpropagation with semantic gradients -- a generalization that aligns several key optimization techniques, including reverse-mode automatic differentiation and the more recent TextGrad by exploiting the relationship among nodes with a common successor. This serves as a method for computing directional information about how changes to each component of an agentic system might improve the system's output. To use these gradients, we propose a method called semantic gradient descent which enables us to solve GASO effectively. Our results on both BIG-Bench Hard and GSM8K show that our approach outperforms existing state-of-the-art methods for solving GASO problems. A detailed ablation study on the LIAR dataset demonstrates the parsimonious nature of our method. A full copy of our implementation is publicly available at https://github.com/HishamAlyahya/semantic_backprop

cs.AI↗