Search arXiv⌕ Search

arXiv subjects

Hong Xu

Publications and source records attributed to Hong Xu.

At least 19 recordsLinked to original sources
arXiv2026

RoutingBench: Can Agentic Routing Analysis Scale to Production Datacenter Networks?

Recent advances in AI models and agentic technologies make AI for network operations (NetOps) within reach. However, scalability remains a key bottleneck of agentic NetOps when analyzing hyperscale networks, which comprise hundreds of datacenters, each housing thousands of network devices. The scalability challenge is rooted in the requirement of many NetOps tasks that must conduct global reasoning on how a local change of device behavior affects all relevant routing paths, known as routing-path analysis. This paper studies this scalability problem and evaluates how different agentic approaches, namely in-context learning, iterative reasoning, and agent skills, can scale routing-path analysis to large, complex networks. We present RoutingBench for evaluating agentic routing-path analysis, with varying network size and complexity, for various types of device changes. Our results show that agentic analysis is promising: agent skills curated with a principle termed "explore more; digest less" enables routing-path analysis on hyperscale networks of 50K routers with an accuracy of 99.5%, significantly outreaching the scalability of traditional symbolic analysis. Meanwhile, RoutingBench also reveals the boundary of AI agent capability on complex inter-datacenter networks and compound changes, posing open challenges for AI and agentic research.

cs.NI↗
arXiv2026

Libra: Efficient Resource Management for Agentic RL Post-Training

Reinforcement learning (RL) has emerged as a standard post-training paradigm for shaping large language models (LLMs) into capable agents. In agentic RL, the rollout stage generates trajectories while invoking tools, producing long-tailed and non-stationary workloads that expose two fundamental challenges. First, due to the long-tailed response distribution, a small fraction of trajectories dominates rollout makespan.Second, rollout and training differ in their compute patterns, memory demands, and sensitivity to sequence length. As the policy evolves, shifts in the workload distribution further change their relative resource demands, making it difficult to maintain balanced execution across the two stages. We present Libra, an adaptive runtime for agentic RL post-training with two complementary components: (1) intra-stage scheduling via a Causality-Guided Bucket Scheduler that routes requests across execution buckets with different parallelism configurations, reducing delays from rollout stragglers; and (2) cross-stage coordination that dynamically reallocates workers between rollout and training as the workload changes. It moves workers between the two stages through a non-blocking protocol without interrupting ongoing training. Evaluated on a 48x NVIDIA A800 GPU cluster and a 160x Ascend 910B3 NPU cluster across three agentic benchmarks, Libra achieves up to 4.2x higher throughput and up to 2.7x faster reward convergence

cs.LG↗
arXiv2026

Exploiting Structure in the Boolean Weighted Constraint Satisfaction Problem: A Constraint Composite Graph-Based Approach

What is "structure," and how can we exploit it in combinatorial optimization? Reasoning tasks such as system design, state estimation, and prediction can be cast as combinatorial optimization problems (COPs), traditionally attacked by dedicated algorithms that fail outside their narrow subclass, while communities working on equivalent problems reinvent each other's techniques. The weighted constraint satisfaction problem (WCSP) is a general framework that subsumes COPs from many communities and represents complex physical and non-physical systems, from spin glasses to social networks. Is such generality a curse? Our answer is to exploit "structure": a general-purpose WCSP algorithm should automatically imitate the specialized algorithm for whatever subclass its input belongs to. A WCSP has macro (graphical) structure, which variables interact, and micro (numerical) structure, how they interact. Separate schools of thought exploit one or the other, never both. The constraint composite graph (CCG), introduced in 2008, unifies them: it is an undirected graph over a WCSP instance's variables plus auxiliary ones, on which minimum weighted vertex cover (MWVC) solves the original instance. It is efficiently constructible, but largely unexploited. This dissertation answers three questions affirmatively. The CCG has theoretical advantages beyond identifying tractable classes (new properties proved here); it is practically useful (efficient implementation and experiments); and it extends promisingly to non-Boolean variables (new encodings). We use the CCG to (a) kernelize a WCSP instance, fixing optimal values of some variables by maxflow before search begins, (b) improve min-sum message passing, (c) exploit integer linear programming solvers, and (d) solve COPs on quantum annealers. Solving general COPs faster classically also yields better baselines for the debated quantum advantage.

cs.LO↗
arXiv2026

Affix Cache for Diffusion Large Language Models

Diffusion Large Language Models (DLLMs) enable non-autoregressive decoding and bidirectional context modeling, but efficient inference remains challenging. Unlike autoregressive systems, whose key-value (KV) cache can be reused for shared prefixes, DLLMs couple the KV states of shared context tokens with evolving generated tokens through bidirectional attention, making naive cache reuse stale while full recomputation is expensive. We present ACache, an affix-oriented cache reuse mechanism for shared text spans in DLLMs beyond prefixes. ACache identifies a small request-specific subset of critical affix tokens, called Anchor Tokens, by measuring their influence on masked generation tokens, and selectively recomputes the KV states of only these tokens while reusing the remaining affix cache. Built on Fast-dLLM, ACache recovers the accuracy loss caused by direct affix-cache reuse across different settings when recomputing around 20% of affix tokens. We also build a shared-prefix prototype on top of the Nano-vLLM engine, showing that ACache reduces recompute latency by up to 55.7% and improves end-to-end throughput by up to 1.68$\times$.

cs.CL↗
arXiv2026

Transferable Attack against Face Swapping in an Extended Space

Although deep Face Swapping (FS) models may benefit the entertainment industry, they pose severe threats to privacy and security. Existing protections, including deepfake detection and adversarial perturbation, are either passive responses or ineffective to unseen subject-agnostic FS models. In this paper, we propose a transferable attack against subject-agnostic FS models named Additive Identity attack based on a Relighting function (AIR). AIR leverages reillumination and additive perturbations to mislead the identity extraction modules in subject-agnostic FS models. By using these two types of perturbations simultaneously, the attack space is extended such that stronger but more visually natural adversarial examples can be identified. To further enhance the visual quality while preserving the effectiveness of the attack, an adaptive translation-invariant operation and an illumination control scheme are designed for AIR. Unlike other methods, AIR does not require a surrogate FS model to achieve high transferability. In addition, a mathematical proof is given for the extension of the attack space. Extensive experiments using 1000 image pairs across various state-of-the-art subject-agnostic FS models, including GAN and diffusion-based FS models, show that AIR surpasses all existing attacks in terms of both attack success rate and image quality.

cs.CV↗
arXiv2026

Frontier: Towards Comprehensive and Accurate LLM Inference Simulation

Modern LLM serving is no longer homogeneous or monolithic. Production systems now combine disaggregated execution, complex parallelism, runtime optimizations, and stateful workloads such as reasoning, agents, and RL rollouts. Simulation is attractive for exploring this growing design space, yet existing simulators lack the architectural completeness and decision-grade fidelity it demands. Their monolithic-replica abstractions are ill-suited to disaggregated serving, while average-case analytical proxies can distort SLA predictions and even reverse optimization conclusions. We present Frontier, a discrete-event simulator for modern LLM inference serving. Frontier features a disaggregated abstraction. It captures the structure and dynamics of modern serving systems by modeling co-location, Prefill-Decode Disaggregation (PDD), and Attention-FFN Disaggregation (AFD) with role-specific cluster workers, incorporating key runtime optimizations (e.g., CUDA Graphs, speculative decoding) within the scheduler-batch-engine loop, and supporting stateful requests for emerging workloads. It further provides accurate and generalizable predictions of computation, communication, and memory costs across diverse serving scenarios with complex workload compositions. On 16-H800 GPU testbed, Frontier achieves an average throughput error below 4%. Compared with state-of-the-art simulators, it reduces end-to-end latency error from 44.9% to 6.4% under co-location and from 51.7% to 2.6% under disaggregation. It scales to over 1K GPUs on commodity CPUs and enables new use cases such as SLA-dependent Pareto frontier exploration, heterogeneous disaggregated allocation, agentic reasoning scheduling validation, and RL post-training reconfiguration. We release Frontier at https://github.com/NetX-lab/Frontier.

cs.DC↗
arXiv2026

Hydrogen Development in China and the EU: A Recommended Tian Ji's Horse Racing Strategy

The global momentum towards establishing sustainable energy systems has become increasingly prominent. Hydrogen, as a remarkable carbon-free and renewable energy carrier, has been endorsed by 39 countries at COP28 in the UAE, recognizing its essential role in global energy transition and industry decarbonization. Both the European Union (EU) and China are at the forefront of this shift, developing hydrogen strategies to enhance regional energy security and racing for carbon neutrality commitments by 2050 for the EU and 2060 for China. The wide applications of hydrogen across hard-to-abate sectors and the flexibility of decentralized production and storage offer customized solutions utilizing local resources in a self-paced manner. To unveil the trajectory of hydrogen development in China and the EU, this paper proposes a comparative analysis framework employing key factors to investigate hydrogen developments in both economic powerhouses. Beyond country-wise statistics, it dives into representative hydrogen economic areas in China (Inner Mongolia, Capital Economic Circle, Yangtze River Delta) and Europe (Delta Rhine Corridor) for understanding supply and demand, industrial synergy, and policy incentives for local hydrogen industries. The derived implications offer stakeholders an evolving hydrogen landscape across the Eurasian continent and insights for future policy developments facilitating the global green transition.

q-fin.GN↗
arXiv2026

FOCUS: DLLMs Know How to Tame Their Compute Bound

Diffusion Large Language Models (DLLMs) offer a compelling alternative to Auto-Regressive models, but their deployment is constrained by high decoding cost. In this work, we identify a key inefficiency in DLLM decoding: while computation is parallelized over token blocks, only a small subset of tokens is decodable at each diffusion step, causing most compute to be wasted on non-decodable tokens. We further observe a strong correlation between attention-derived token importance and token-wise decoding probability. Based on this insight, we propose FOCUS, an inference system designed for DLLMs. By dynamically focusing computation on decodable tokens and evicting non-decodable ones on-the-fly, FOCUS increases the effective batch size, alleviating compute limitations and enabling scalable throughput. Empirical evaluations demonstrate that FOCUS achieves up to 3.52$\times$ throughput improvement over the production-grade engine LMDeploy in large-batch settings, while preserving or improving generation quality across multiple benchmarks.

cs.LG↗
arXiv2026

UniScale: Adaptive Unified Inference Scaling via Online Joint Optimization of Model Routing and Test-Time Scaling

In real-world deployments of large language models (LLMs), balancing inference quality and computational cost has become a central challenge. Existing approaches tackle this trade-off along two largely independent dimensions: model routing, which switches among models of different scales to match request complexity, and test-time scaling (TTS), which adjusts inference-time compute within a fixed model for fine-grained control. However, this decoupled design introduces inherent limitations. Model routing yields coarse-grained, discrete performance changes due to the sparse set of model scales, while single-model TTS often encounters capacity ceilings and exhibits diminishing returns as compute increases. Moreover, treating the two mechanisms separately restricts adaptability in dynamic inference environments. To overcome these limitations, we introduce Unified Inference Scaling (UIS), which unifies model routing and TTS in a single optimization space. Building on this formulation, we propose UniScale, an online framework that models adaptive UIS as a contextual multi-armed bandit problem and learns inference policies via LinUCB. The framework incorporates efficiency-aware learning and cost modeling to ensure stable and scalable optimization over high-dimensional action spaces. Evaluation shows that UniScale effectively exploits the synergy in the UIS space to deliver a fine-grained and consistently better quality-cost trade-off across diverse, dynamic inference scenarios.

cs.AI↗
arXiv2026

Towards Reliable Fetal Ultrasound Interpretation with Multi-Agent Collaboration

Automated fetal ultrasound interpretation requires a workflow from visual perception, including plane recognition and anatomical segmentation, to clinical understanding, including biometric measurement and diagnostic reporting. However, the prevailing "one-task, one-model" paradigm limits systematic integration of evidence across this multi-step process. Although multimodal large language models (MLLMs) show promising visual understanding, their limited domain-specific grounding and hallucination risks restrict reliability in fetal ultrasound analysis. To address these limitations, we propose FetUSAgents, a tool-augmented multi-agent system for comprehensive fetal ultrasound interpretation, supporting visual question answering (VQA), report generation, image captioning, and video summarization. FetUSAgents coordinates task-specific visual tools through collaborative LLM agents and decomposes clinical queries into subtasks that progress from anatomical recognition to quantitative measurement. We further introduce Dual-Path Evidence Arbitration (DPEA), which integrates LLM-based deliberative reasoning with structured computational evidence from specialized visual tools. A retrieval-enhanced evidence bank consolidates intermediate findings to support traceable and clinically grounded conclusions. In addition, we construct FetUS-VQA, a dedicated VQA benchmark for fetal ultrasound, comprising 1,892 images and 3,205 question-answer pairs across 10 clinical tasks. Extensive out-of-distribution experiments show that FetUSAgents outperforms general and medical MLLMs, exceeding the strongest baseline by more than 25 percent in VQA accuracy. These results suggest a scalable route toward evidence-driven clinical assistants for prenatal imaging. Code is available.

cs.CV↗
arXiv2026

PolyGnosis 2.0: Enhancing LLM Reasoning via Agentic Harness Engineering for Polymarket and OSINT Insight Extraction

This paper introduces PolyGnosis 2.0, a pioneering multi-agent architecture designed to extract predictive intelligence by synthesizing Polymarket anomaly signals with global Open Source Intelligence (OSINT) streams, specifically Global Database of Events, Language, and Tone (GDELT). We define and target "Perspective Mismatches", the narrative divergence between Polymarket sentiment and global media flows, as high-alpha trading signals. Moving beyond generic agentic superiority, we rigorously quantify the efficacy of "Harness Engineering" techniques, including reflection loops, tool-calling, divide-and-conquer partitioning (D&C), and chain-of-thought (CoT), within high-noise financial domains. Our empirical evaluation against human-expert benchmarks reveals that while structural partitioning is mandatory for multi-dimensional alignment, unconstrained terminal reflection actively induces logical drift. Furthermore, we identify a pervasive "consensus bias" across all agent configurations during narrative reasoning, necessitating deterministic validation. Ultimately, we isolate a Pareto-optimal configuration that achieves professional-grade analytical precision while minimizing latency and token overhead, providing a robust blueprint for autonomous intelligence in prediction markets.

cs.CL↗
arXiv2026

AGZO: Activation-Guided Zeroth-Order Optimization for LLM Fine-Tuning

Zeroth-Order (ZO) optimization has emerged as a promising solution for fine-tuning LLMs under strict memory constraints, as it avoids the prohibitive memory cost of storing activations for backpropagation. However, existing ZO methods typically employ isotropic perturbations, neglecting the rich structural information available during the forward pass. In this paper, we identify a crucial link between gradient formation and activation structure: the gradient of a linear layer is confined to the subspace spanned by its input activations. Leveraging this insight, we propose Activation-Guided Zeroth-Order optimization (AGZO). Unlike prior methods, AGZO extracts a compact, activation-informed subspace on the fly during the forward pass and restricts perturbations to this low-rank subspace. We provide a theoretical framework showing that AGZO optimizes a subspace-smoothed objective and provably yields update directions with higher cosine similarity to the true gradient than isotropic baselines. Empirically, we evaluate AGZO on Qwen3 and Pangu models across various benchmarks. AGZO consistently outperforms state-of-the-art ZO baselines and significantly narrows the performance gap with first-order fine-tuning, while maintaining almost the same peak memory footprint as other ZO methods.

cs.LG↗
arXiv2026

Defining Robust Ultrasound Quality Metrics via an Ultrasound Foundation Model

Clinicians lack a principled framework to quantify diagnostic utility in ultrasound reconstructions. Existing standards like PSNR and VGG-LPIPS are inadequate, failing to account for modality-specific physics or the structural nuances of acoustic imaging. We close this gap with a TinyUSFM-based evaluation framework featuring two distinct metrics: TinyUSFM-uLPIPS, a full-reference perceptual distance based on multi-layer token relations, and TinyUSFM-NRQ, a deployable no-reference quality score utilizing clean-manifold modeling and worst-region aggregation to detect localized harmful artifacts. We demonstrate that the presented metrics have four unique advantages: 1) Task-linked quality, where TinyUSFM-uLPIPS achieves superior calibration with semantic task damage, accurately reflecting Dice-score drops in segmentation where VGG-based metrics fail; 2) Cross-organ comparability, maintaining stable scoring scales and consistent severity rankings across diverse anatomical sites and domain-shifted data; 3) PSNR-consistent sensitivity, with TinyUSFM-NRQ providing a reliable quality score without ground-truth images that remains consistent with traditional fidelity benchmarks (i.e. PSNR); and 4) Clinical utility, improving the prediction of expert preference from 47.2$\%$ to 72.8$\%$ accuracy and producing super-resolution reconstructions preferred by sonographers. By integrating these advantages into a unified assessment and optimization loop, this work establishes a modality-aligned standard that finally bridges the gap between algorithmic performance and diagnostic utility. Our code is available at https://github.com/sextant-fable/US-Metrics.

eess.IV↗
arXiv2026

ReasonCache: Accelerating Large Reasoning Model Serving through KV Cache Sharing

Large Reasoning Models (LRMs) are becoming integral to many AI inference systems, enhancing their capabilities with advanced reasoning. However, deploying these models in production environments presents a significant QoS challenge: the substantial memory overhead from their long, auto-regressive inference processes severely limits throughput and increases latency, thereby affecting the quality of service for concurrent users. We observe that LRMs frequently generate highly similar intermediate reasoning steps, which, in turn, correspond to highly similar KV cache states across layers. Building on this insight, we propose ReasonCache, a novel KV cache management approach designed to improve the QoS of AI inference systems. ReasonCache utilizes a Collaborative Filtering Algorithm to efficiently identify reusable KV cache blocks and enables zero-copy cache reuse. Experimental evaluation demonstrates that ReasonCache achieves a peak throughput improvement of 89.2% and an average gain of 40-60%, leading to more responsive and cost-effective AI inference services. Notably, this performance is achieved while maintaining higher accuracy compared to existing KV cache management techniques.

cs.LG↗
arXiv2026

GameGen-Verifier: Parallel Keypoint-Based Verification for LLM-Generated Games via Runtime State Injection

LLM-based game generation promises to turn natural-language specifications into executable games, but progress is limited by the lack of reliable automated verification. Unlike conventional code generation, game correctness is defined over long-horizon interaction: a game may appear correct while violating core mechanics such as state updates, interaction rules, and phase transitions. Existing Agent-as-a-Verifier approaches collapse verification into open-ended gameplay, making verdicts reachability-bound, time-consuming, coverage-limited, and sensitive to the agent's gameplay ability. We present GameGen-Verifier, an automated verification paradigm for LLM-generated games that decomposes a specification into verifiable keypoints and grounds them into independent verification units. Each unit patches the game runtime into a concrete target state, executes a bounded interaction, and judges the outcome against the keypoint assertion. We implement GGV-Harness, a scalable agentic harness providing concurrency management, runtime isolation, and fault recovery. On VeriGame, our dataset of 100 games across seven genres, GameGen-Verifier achieves up to 92.2% accuracy against human judgments versus 58.8% for the coverage-enforced Agent-as-a-Verifier baseline, while reducing wall-clock time by up to 16.6x.

cs.LG↗
arXiv2026

TSGuard: Automated User-Centric Incident Diagnosis for AI Workloads in the Cloud

AI workloads incur frequent failures and incidents from the underlying infrastructure. The current incident management workflow follows a provider-centric paradigm, where users report incidents to the infrastructure provider who then conducts troubleshooting. Due to the large number of incidents and the manual nature of the troubleshooting process, the provider often takes several days to resolve an incident, resulting in operational delays and productivity loss. To address these challenges, we present TSGuard, a user-centric multi-agent system that delivers immediate incident diagnosis to users who deploy the workloads. The core innovation of TSGuard is twofold: (1) constructing domain-specific knowledge bases by mining historical on-call experiences in the offline phase, and (2) mimicking human expert diagnosis via structured reasoning and iterative trial-and-error in the online phase. Evaluation using production incident records from Microsoft Azure demonstrates that TSGuard significantly outperforms state-of-the-art baselines, improving diagnostic accuracy by 19.8%. Furthermore, TSGuard reduces the average verification time by 63.4% compared to the sequential execution baseline.

cs.SE↗
arXiv2026

Piezomagnetic Switching of Nonvolatile Antiferromagnetic States

Prospective spintronic memory and logic devices will benefit from the negligible stray field and ultrafast magnetic dynamics inherent to antiferromagnets [1]. However, realizing isothermal, nonvolatile, and deterministic switching of antiferromagnetic states remains a key challenge [2, 3]. Here, we propose a piezomagnetic writing scheme in triangular Mn3Ir-based memory cells, with readout achieved via the exchange bias effect. Our approach enables deterministic and nonvolatile switching of the antiferromagnetic states, which exhibit exceptional robustness against external perturbations. The switching mechanism is ascribed to piezomagnetic effect of Mn3Ir combined with the interfacial Dzyaloshinskii-Moriya interaction at the antiferromagnet-ferromagnet interface. This scheme overcomes the speed limitations imposed by conventional isothermal methods based on isothermal crystallization mechanism [4]. Our findings highlight the potential of piezomagnetic effects in designing advanced spintronic devices, providing an efficient pathway for manipulating antiferromagnetic states and developing energy-efficient memory technology.

cond-mat.mtrl-sci↗
arXiv2026

A Systematic Evaluation of On-Device LLMs: Quantization, Performance, and Resources

Deploying Large Language Models (LLMs) on edge devices enhances privacy but faces performance hurdles due to limited resources. We introduce a systematic methodology to evaluate on-device LLMs, balancing capability, efficiency, and resource constraints. Through an extensive analysis of models (0.5B-14B) and seven post-training quantization (PTQ) methods on commodity hardware, we demonstrate that: 1) Heavily quantized large models consistently outperform smaller, high-precision models, with a performance threshold at ~3.5 effective bits-per-weight (BPW); 2) Resource utilization scales linearly with BPW, though power and memory footprints vary by quantization algorithm; and 3) With a reduction in model size, the primary constraint on throughput transitions from communication overhead to computational latency. We conclude by offering guidelines for optimizing LLMs in resource-constrained edge environments. Our codebase is available at https://anonymous.4open.science/r/LLMOnDevice/.

cs.LG↗