arXiv subjects

Chenxi Yang

Publications and source records attributed to Chenxi Yang.

At least 19 recordsLinked to original sources

arXiv2026

ComBodied Agents: a New Paradigm of Human-Centric Agentic AI

After an older adult misses a medication dose, a software agent can send another reminder and an embodied agent can bring the medication. Yet neither explains whether the person forgot, is confused, has side effects, or deliberately refused, nor what support is appropriate. This reveals a structural gap in Agentic AI: Digital Agents primarily transform software states, while Embodied Agents transform physical states; neither makes a person's evolving state and agency the primary object of modeling, intervention, and evaluation. We introduce Combodied Agents, a human-centered paradigm that perceives, models, predicts, and supports individual human-state trajectories over time, using software tools, sensors, wearables, robots, and human services as action channels rather than end goals. We unify fragmented capabilities across personal assistants, health agents, AI companions, and adaptive human--AI systems into a closed loop: event-based multimodal perception reconstructs meaningful personal events; longitudinal, correctable memory provides temporal context; Personal World Models estimate future personal states and outcomes under alternative decisions and interventions; and an admissible intervention policy selects proportionate support under consent, uncertainty, safety, reversibility, and user control. Feedback from the person and environment updates the loop. Rather than requiring an exhaustive Human Digital Twin, the framework uses purpose-bounded, uncertainty-aware, user-correctable representations. We organize the design space by human-state targets, relational contexts, and agent roles, and propose scenario-centered evaluation, agency-preservation metrics, benchmark requirements, edge-native personal models, and governance directions. Combodied Agents shift Agentic AI from external task completion toward sustained human benefit.

cs.AI↗

arXiv2026

Diagnosing as Cardiologists Do: ECG Agents with Doctor-Grounded Priors for Clinical Reasoning Across Diseases and Populations

Cardiologists interpret electrocardiograms by localizing waveform components, measuring rhythm and interval patterns, and translating these structured observations into diagnostic evidence. Whether this expert reading process can serve as an effective prior for ECG agents remains unclear. To address this question, we introduce LuminaECG, a clinically structured ECG reasoning framework that reformulates ECG interpretation as measurement-grounded visual reading. ECG signals are rendered on standard electrocardiographic grid paper to preserve the spatial and scale cues used in clinical reading. P-wave, QRS-complex, and T-wave boundaries are explicitly delineated, and color-coded segmentation decomposes the waveform into discrete visual measurement primitives. A general 2B vision-language backbone is then trained with low-rank supervised fine-tuning to associate these primitives with diagnostic reasoning, without architectural modification. Across open, proprietary, and ECG-specialist zero-shot baselines, LuminaECG improves both waveform measurement and diagnostic recovery. It reaches a clinically meaningful reader tier on the CODE-test benchmark, transfers across geographically diverse ECG datasets without retraining, and generates reports whose structure contains an emergent prognostic signal. These findings suggest that effective ECG agents require not only larger models, but supervision that preserves the alignment between measurable waveform evidence and clinical knowledge.

eess.SP↗

arXiv2026

A Sharp Ramsey Theorem for Admissible Colorings of Ordered Cliques

Let $f(k)$ be the minimum integer $N$ such that any red--blue edge-coloring of the ordered complete graph on $N$ vertices contains a set of $k$ vertices whose induced coloring is admissible. In this note, we obtain the exact value of $f(k)$ for $k\ge 3$, which confirms a conjecture posed by Brada\v{c}, Liu, Wu and Xu.

math.CO↗

arXiv2026

Three Results on Generalized Quasikernels in Digraphs

A $q$-kernel of a digraph $D$ is an independent set $Q\subseteq D$ such that every vertex of $D$ is reachable from $Q$ by a directed path of length at most $q$, which is a natural generalization of kernels and quasikernels. In this paper, we establish three results on generalized quasikernels. Firstly, we prove that any $n$-vertex source-free bipartite oriented graph with no directed 4-cycles has a quasikernel of size at most $17n/35$. Secondly, we show that every digraph with no $(r-1)$-source set contains $r$ pairwise disjoint $(3r-2)$-kernels, where $r\ge 2$. At last, we consider unicyclic digraph with a directed cycle of length $2\ell$ and bipartition $U\cup V$, and we prove that for every odd integer $q\ge 3$, there exist two $q$-kernels $Q_U\subseteq U$ and $Q_V\subseteq V$ such that \[ |Q_U|+|Q_V| \le 2\cdot \frac{\lceil \ell/(q+1)\rceil}{\ell} |V(D)|. \] These results confirm two conjectures and give an affirmative answer to a question posed by Spiro in European Journal of Combinatorics 133 (2026), 104307.

math.CO↗

arXiv2026

Connected graphs with a large dissociation number attaining the minimum spectral radius

A dissociation set in a graph is a subset of vertices that induces a subgraph of maximum degree at most one, which is a natural generalization of the notion of an independent set. The dissociation number of a graph is defined as the maximum cardinality of a dissociation set. This paper studies the minimum spectral radius of connected graphs with a given order $n$ and a given dissociation number $\psi$. For $\psi=n-k$ with $k\ge 4$ fixed and $n$ sufficiently large, we establish both upper and lower bounds for this minimum spectral radius and prove the extremal graphs must belong to a specific graph class.

math.CO↗

arXiv2026

Towards Demystifying and Repairing LLM-in-the-Loop Vulnerabilities

Large Language Models (LLMs) have been actively integrated into modern software systems as critical components, introducing a new type of software vulnerability, LLM-in-the-Loop (LiL) vulnerability, in which threats are caused by LLMs. Although some studies have attempted to investigate the impact of LiL vulnerabilities, they have unfortunately failed to clearly distinguish LiL vulnerabilities from conventional ones, leaving the understanding of real-world LiL vulnerabilities an open problem. To address this gap, we first clearly define the scope of LiL vulnerability, and discuss the differences between LiL vulnerabilities and vulnerabilities that exist in LLM systems but are not really caused by LLMs (i.e., LLM-ecosystem vulnerabilities). Then, we construct the first LiL vulnerability dataset, LiLCVE, covering 41 LiL vulnerabilities and 75 LLM-ecosystem vulnerabilities, to facilitate the risk analysis of LLM-integrated software. The analysis of LiLCVE reveals that LiL vulnerabilities have higher severity than LLM-ecosystem vulnerabilities and conventional software vulnerabilities, with 15.5% and 30.3% more critical vulnerabilities, respectively. Furthermore, given the high severity of LiL vulnerabilities and the potential of LLM-based vulnerability repair methods in patching conventional software vulnerabilities. We explore the capabilities of existing widely-used LLM-based methods in repairing vulnerabilities in LiLCVE. Experimental results on 20 agent-model configuration demonstrate that LiL vulnerabilities are far more challenging to fix, with an average decrease of 10.8% Pass@1 rate compared to other types of vulnerabilities. More critically, three categories, Generated Query Execution, Agent Action, and Model Output Rendering, frequently receive 0% repair success rates.

cs.SE↗

arXiv2026

Connected graphs minimizing the spectral radius for given order and dissociation number

A dissociation set in a graph is a subset of vertices which induces a subgraph with maximum degree at most one. The dissociation number of a graph is the maximum cardinality of its dissociation sets. In this paper, we consider the $n$-vertex connected graphs with a given dissociation number that attain the minimum spectral radius. By using structure analysis and constructing difference equations, we characterize the extremal graphs with dissociation number $n-3$.

math.CO↗

arXiv2026

Observation of anomalous exciton polariton bands in PEPI perovskite based microcavity at room temperature

Recently anomalous energy bands with negative mass attract intensive attention where non Hermiticity plays an important role. In this work we observe anomalous exciton polariton bands in PEPI perovskite based microcavity at room temperature. We simulate the anomalous band structure using a non-Hermitian coupled oscillator model which agree with experiments very well. Our results offer to study non-Hermitian polariton wave dynamics at room temperature.

physics.optics↗

arXiv2025

Taming a Retrieval Framework to Read Images in Humanlike Manner for Augmenting Generation of MLLMs

Multimodal large language models (MLLMs) often fail in fine-grained visual question answering, producing hallucinations about object identities, positions, and relations because textual queries are not explicitly anchored to visual referents. Retrieval-augmented generation (RAG) alleviates some errors, but it fails to align with human-like processing at both the retrieval and augmentation levels. Specifically, it focuses only on global-level image information but lacks local detail and limits reasoning about fine-grained interactions. To overcome this limitation, we present Human-Like Retrieval-Augmented Generation (HuLiRAG), a framework that stages multimodal reasoning as a ``what--where--reweight'' cascade. Queries are first anchored to candidate referents via open-vocabulary detection (what), then spatially resolved with SAM-derived masks to recover fine-grained precision (where), and adaptively prioritized through the trade-off between local and global alignment (reweight). Mask-guided fine-tuning further injects spatial evidence into the generation process, transforming grounding from a passive bias into an explicit constraint on answer formulation. Extensive experiments demonstrate that this human-like cascade improves grounding fidelity and factual consistency while reducing hallucinations, advancing multimodal question answering toward trustworthy reasoning.

cs.CV↗

arXiv2025

A Bring-Your-Own-Model Approach for ML-Driven Storage Placement in Warehouse-Scale Computers

Storage systems account for a major portion of the total cost of ownership (TCO) of warehouse-scale computers, and thus have a major impact on the overall system's efficiency. Machine learning (ML)-based methods for solving key problems in storage system efficiency, such as data placement, have shown significant promise. However, there are few known practical deployments of such methods. Studying this problem in the context of real-world hyperscale data centers at Google, we identify a number of challenges that we believe cause this lack of practical adoption. Specifically, prior work assumes a monolithic model that resides entirely within the storage layer, an unrealistic assumption in real-world deployments with frequently changing workloads. To address this problem, we introduce a cross-layer approach where workloads instead ''bring their own model''. This strategy moves ML out of the storage system and instead allows each workload to train its own lightweight model at the application layer, capturing the workload's specific characteristics. These small, interpretable models generate predictions that guide a co-designed scheduling heuristic at the storage layer, enabling adaptation to diverse online environments. We build a proof-of-concept of this approach in a production distributed computation framework at Google. Evaluations in a test deployment and large-scale simulation studies using production traces show improvements of as much as 3.47$\times$ in TCO savings compared to state-of-the-art baselines.

cs.DC↗

arXiv2024

Canopy: Property-Driven Learning for Congestion Control

Learning-based congestion controllers offer better adaptability compared to traditional heuristics. However, the unreliability of learning techniques can cause learning-based controllers to behave poorly, creating a need for formal guarantees. While methods for formally verifying learned congestion controllers exist, these methods offer binary feedback that cannot optimize the controller toward better behavior. We improve this state-of-the-art via Canopy, a new property-driven framework that integrates learning with formal reasoning in the learning loop. Canopy uses novel quantitative certification with an abstract interpreter to guide the training process, rewarding models, and evaluating robust and safe model performance on worst-case inputs. Our evaluation demonstrates that unlike state-of-the-art learned controllers, Canopy-trained controllers provide both adaptability and worst-case reliability across a range of network conditions.

cs.LG↗

arXiv2024

Nonregular graphs with a given maximum degree attaining maximum spectral radius

Let $G$ be a connected nonregular graphs of order $n$ with maximum degree $\Delta$ that attains the maximum spectral radius. Liu and Li (2008) proposed a conjecture stating that $G$ has a degree sequence $(\Delta,\ldots,\Delta,\delta)$ with $\delta<\Delta$. For $\Delta=3$ and $\Delta=4$, Liu (2024) confirmed this conjecture by characterizing the structure of such graphs. Liu also proposed a modified version of the conjecture for fixed $\Delta$ and sufficiently large $n$, stating that the above $\delta=\Delta-1$ if $\Delta$ and $n$ are both odd, $\delta=1$ if $\Delta$ is odd and $n$ is even, and $\delta=\Delta-2$ if $\Delta$ is even. For the cases where $\Delta=n-2$ with $n\ge 5$, and $\Delta=n-3$ with $n\ge 59$, we fully characterize the structure of $G$.

math.CO↗

arXiv2024

Extremal oriented graphs avoiding 1-subdivision of an in-star

An oriented graph is a digraph obtained from an undirected graph by choosing an orientation for each edge. Given a positive integer $n$ and an oriented graph $F$, the oriented Tur$\acute{\rm a}$n number $ex_{ori}(n,F)$ is the maximum number of arcs in an $F$-free oriented graph of order $n$. In this paper, we investigate the oriented Tur$\acute{\rm a}$n number $ex_{ori}(n, \overrightarrow{S_{k,1}} )$, where $\overrightarrow{S_{k,1}}$ is the $1$-subdivision of the in-star of order $k+1$. We determine $ex_{ori}(n,\overrightarrow{S_{k,1}}) $ for $k=2,3$ as well as the extremal oriented graphs. For $k\ge 4$, we establish a lower bound and an upper bound on $ex_{ori}(n,\overrightarrow{S_{k,1}})$.

math.CO↗

arXiv2024

Exploring Vulnerabilities of No-Reference Image Quality Assessment Models: A Query-Based Black-Box Method

No-Reference Image Quality Assessment (NR-IQA) aims to predict image quality scores consistent with human perception without relying on pristine reference images, serving as a crucial component in various visual tasks. Ensuring the robustness of NR-IQA methods is vital for reliable comparisons of different image processing techniques and consistent user experiences in recommendations. The attack methods for NR-IQA provide a powerful instrument to test the robustness of NR-IQA. However, current attack methods of NR-IQA heavily rely on the gradient of the NR-IQA model, leading to limitations when the gradient information is unavailable. In this paper, we present a pioneering query-based black box attack against NR-IQA methods. We propose the concept of score boundary and leverage an adaptive iterative approach with multiple score boundaries. Meanwhile, the initial attack directions are also designed to leverage the characteristics of the Human Visual System (HVS). Experiments show our method outperforms all compared state-of-the-art attack methods and is far ahead of previous black-box methods. The effective NR-IQA model DBCNN suffers a Spearman's rank-order correlation coefficient (SROCC) decline of 0.6381 attacked by our method, revealing the vulnerability of NR-IQA models to black-box attacks. The proposed attack method also provides a potent tool for further exploration into NR-IQA robustness.

cs.CV↗

arXiv2024

Beyond Score Changes: Adversarial Attack on No-Reference Image Quality Assessment from Two Perspectives

Deep neural networks have demonstrated impressive success in No-Reference Image Quality Assessment (NR-IQA). However, recent researches highlight the vulnerability of NR-IQA models to subtle adversarial perturbations, leading to inconsistencies between model predictions and subjective ratings. Current adversarial attacks, however, focus on perturbing predicted scores of individual images, neglecting the crucial aspect of inter-score correlation relationships within an entire image set. Meanwhile, it is important to note that the correlation, like ranking correlation, plays a significant role in NR-IQA tasks. To comprehensively explore the robustness of NR-IQA models, we introduce a new framework of correlation-error-based attacks that perturb both the correlation within an image set and score changes on individual images. Our research primarily focuses on ranking-related correlation metrics like Spearman's Rank-Order Correlation Coefficient (SROCC) and prediction error-related metrics like Mean Squared Error (MSE). As an instantiation, we propose a practical two-stage SROCC-MSE-Attack (SMA) that initially optimizes target attack scores for the entire image set and then generates adversarial examples guided by these scores. Experimental results demonstrate that our SMA method not only significantly disrupts the SROCC to negative values but also maintains a considerable change in the scores of individual images. Meanwhile, it exhibits state-of-the-art performance across metrics with different categories. Our method provides a new perspective on the robustness of NR-IQA models.

eess.IV↗

arXiv2024

LTL-Constrained Policy Optimization with Cycle Experience Replay

Linear Temporal Logic (LTL) offers a precise means for constraining the behavior of reinforcement learning agents. However, in many settings where both satisfaction and optimality conditions are present, LTL is insufficient to capture both. Instead, LTL-constrained policy optimization, where the goal is to optimize a scalar reward under LTL constraints, is needed. This constrained optimization problem proves difficult in deep Reinforcement Learning (DRL) settings, where learned policies often ignore the LTL constraint due to the sparse nature of LTL satisfaction. To alleviate the sparsity issue, we introduce Cycle Experience Replay (CyclER), a novel reward shaping technique that exploits the underlying structure of the LTL constraint to guide a policy towards satisfaction by encouraging partial behaviors compliant with the constraint. We provide a theoretical guarantee that optimizing CyclER will achieve policies that satisfy the LTL constraint with near-optimal probability. We evaluate CyclER in three continuous control domains. Our experimental results show that optimizing CyclER in tandem with the existing scalar reward outperforms existing reward-shaping methods at finding performant LTL-satisfying policies.

cs.LG↗

arXiv2024

Defense Against Adversarial Attacks on No-Reference Image Quality Models with Gradient Norm Regularization

The task of No-Reference Image Quality Assessment (NR-IQA) is to estimate the quality score of an input image without additional information. NR-IQA models play a crucial role in the media industry, aiding in performance evaluation and optimization guidance. However, these models are found to be vulnerable to adversarial attacks, which introduce imperceptible perturbations to input images, resulting in significant changes in predicted scores. In this paper, we propose a defense method to improve the stability in predicted scores when attacked by small perturbations, thus enhancing the adversarial robustness of NR-IQA models. To be specific, we present theoretical evidence showing that the magnitude of score changes is related to the $\ell_1$ norm of the model's gradient with respect to the input image. Building upon this theoretical foundation, we propose a norm regularization training strategy aimed at reducing the $\ell_1$ norm of the gradient, thereby boosting the robustness of NR-IQA models. Experiments conducted on four NR-IQA baseline models demonstrate the effectiveness of our strategy in reducing score changes in the presence of adversarial attacks. To the best of our knowledge, this work marks the first attempt to defend against adversarial attacks on NR-IQA models. Our study offers valuable insights into the adversarial robustness of NR-IQA models and provides a foundation for future research in this area.

cs.CV↗

arXiv2023

On a Foundation Model for Operating Systems

This paper lays down the research agenda for a domain-specific foundation model for operating systems (OSes). Our case for a foundation model revolves around the observations that several OS components such as CPU, memory, and network subsystems are interrelated and that OS traces offer the ideal dataset for a foundation model to grasp the intricacies of diverse OS components and their behavior in varying environments and workloads. We discuss a wide range of possibilities that then arise, from employing foundation models as policy agents to utilizing them as generators and predictors to assist traditional OS control algorithms. Our hope is that this paper spurs further research into OS foundation models and creating the next generation of operating systems for the evolving computing landscape.

cs.OS↗