arXiv subjects

Xin Yang

Publications and source records attributed to Xin Yang.

At least 73 records · Page 4Linked to original sources

arXiv2026

Modality-Agnostic Prompt Learning for Multi-Modal Camouflaged Object Detection

Camouflaged Object Detection (COD) aims to segment objects that blend seamlessly into complex backgrounds, with growing interest in exploiting additional visual modalities to enhance robustness through complementary information. However, most existing approaches generally rely on modality-specific architectures or customized fusion strategies, which limit scalability and cross-modal generalization. To address this, we propose a novel framework that generates modality-agnostic multi-modal prompts for the Segment Anything Model (SAM), enabling parameter-efficient adaptation to arbitrary auxiliary modalities and significantly improving overall performance on COD tasks. Specifically, we model multi-modal learning through interactions between a data-driven content domain and a knowledge-driven prompt domain, distilling task-relevant cues into unified prompts for SAM decoding. We further introduce a lightweight Mask Refine Module to calibrate coarse predictions by incorporating fine-grained prompt cues, leading to more accurate camouflaged object boundaries. Extensive experiments on RGB-Depth, RGB-Thermal, and RGB-Polarization benchmarks validate the effectiveness and generalization of our modality-agnostic framework.

cs.CV↗

arXiv2026

Silo-Bench: A Scalable Environment for Evaluating Distributed Coordination in Multi-Agent LLM Systems

Large language models are increasingly deployed in multi-agent systems to overcome context limitations by distributing information across agents. Yet whether agents can reliably compute with distributed information, rather than merely exchange it, remains an open question. We introduce SILO-BENCH, a role-agnostic benchmark of 30 algorithmic tasks across three communication complexity levels, evaluating 54 configurations over 1,620 experiments. Our experiments expose a fundamental Communication-Reasoning Gap: agents spontaneously form task-appropriate coordination topologies and exchange information actively, yet systematically fail to synthesize distributed state into correct answers. The failure is localized to the reasoning-integration stage where agents often acquire sufficient information but cannot integrate it. This coordination overhead compounds with scale, eventually eliminating parallelization gains entirely. These findings demonstrate that naively scaling agent count cannot circumvent context limitations, and SILO-BENCH provides a foundation for tracking progress toward genuinely collaborative multi-agent systems. The code is available at https://github.com/jwyjohn/acl26-silo-bench .

cs.MA↗

arXiv2026

A Survey of Continual Reinforcement Learning

Reinforcement Learning (RL) is an important machine learning paradigm for solving sequential decision-making problems. Recent years have witnessed remarkable progress in this field due to the rapid development of deep neural networks. However, the success of RL currently relies on extensive training data and computational resources. In addition, RL's limited ability to generalize across tasks restricts its applicability in dynamic and real-world environments. With the arisen of Continual Learning (CL), Continual Reinforcement Learning (CRL) has emerged as a promising research direction to address these limitations by enabling agents to learn continuously, adapt to new tasks, and retain previously acquired knowledge. In this survey, we provide a comprehensive examination of CRL, focusing on its core concepts, challenges, and methodologies. Firstly, we conduct a detailed review of existing works, organizing and analyzing their metrics, tasks, benchmarks, and scenario settings. Secondly, we propose a new taxonomy of CRL methods, categorizing them into four types from the perspective of knowledge storage and/or transfer. Finally, our analysis highlights the unique challenges of CRL and provides practical insights into future directions.

cs.LG↗

arXiv2026

LogicPoison: Logical Attacks on Graph Retrieval-Augmented Generation

Graph-based Retrieval-Augmented Generation (GraphRAG) enhances the reasoning capabilities of Large Language Models (LLMs) by grounding their responses in structured knowledge graphs. Leveraging community detection and relation filtering techniques, GraphRAG systems demonstrate inherent resistance to traditional RAG attacks, such as text poisoning and prompt injection. However, in this paper, we find that the security of GraphRAG systems fundamentally relies on the topological integrity of the underlying graph, which can be undermined by implicitly corrupting the logical connections, without altering surface-level text semantics. To exploit this vulnerability, we propose \textsc{LogicPoison}, a novel attack framework that targets logical reasoning rather than injecting false contents. Specifically, \textsc{LogicPoison} employs a type-preserving entity swapping mechanism to perturb both global logic hubs for disrupting overall graph connectivity and query-specific reasoning bridges for severing essential multi-hop inference paths. This approach effectively reroutes valid reasoning into dead ends while maintaining surface-level textual plausibility. Comprehensive experiments across multiple benchmarks demonstrate that \textsc{LogicPoison} successfully bypasses GraphRAG's defenses, significantly degrading performance and outperforming state-of-the-art baselines in both effectiveness and stealth. Our code is available at \textcolor{blue}https://github.com/Jord8061/logicPoison.

cs.CL↗

arXiv2026

Diffusion-based Probabilistic Air Quality Forecasting with Mechanistic Insight

Current operational air quality forecasts are computationally expensive, sensitive to errors in physics and emissions, and often neglect weather-related uncertainty. To address these limitations, we present AirFusion, a hybrid, diffusion-based framework that synergistically integrates knowledge from chemical transport models with real-world observational constraints to enable accurate and efficient probabilistic regional air quality prediction. We apply AirFusion to generate operational 6-day, 30-member ensemble forecasts of surface ozone across China, initialized with observations and driven by ensemble weather forecasts. AirFusion outperforms existing operational benchmarks, achieving substantially lower forecast errors against surface measurements, while also providing ensemble-based diagnostics that explicitly quantify the impacts of weather uncertainty on air quality predictability. Moreover, AirFusion can rapidly adapt to evolving emissions through fine-tuning with only one month of recent observations. These attributes establish AirFusion as a powerful and extensible framework for next-generation probabilistic air quality forecasting, with clear potential for application to other pollutants and regions.

physics.ao-ph↗

arXiv2026

PCSTracker: Long-Term Scene Flow Estimation for Point Cloud Sequences

Point cloud scene flow estimation is fundamental to long-term and fine-grained 3D motion analysis. However, existing methods are typically limited to pairwise settings and struggle to maintain temporal consistency over long sequences as geometry evolves, occlusions emerge, and errors accumulate. In this work, we propose PCSTracker, the first end-to-end framework specifically designed for consistent scene flow estimation in point cloud sequences. Specifically, we introduce an iterative geometry motion joint optimization module (IGMO) that explicitly models the temporal evolution of point features to alleviate correspondence inconsistencies caused by dynamic geometric changes. In addition, a spatio-temporal point trajectory update module (STTU) is proposed to leverage broad temporal context to infer plausible positions for occluded points, ensuring coherent motion estimation. To further handle long sequences, we employ an overlapping sliding-window inference strategy that alternates cross-window propagation and in-window refinement, effectively suppressing error accumulation and maintaining stable long-term motion consistency. Extensive experiments on the synthetic PointOdyssey3D and real-world ADT3D datasets show that PCSTracker achieves the best accuracy in long-term scene flow estimation and maintains real-time performance at 32.5 FPS, while demonstrating superior 3D motion understanding compared to RGB-D-based approaches.

cs.CV↗

arXiv2026

Phase-Interface Instance Segmentation as a Visual Sensor for Laboratory Process Monitoring

Reliable visual monitoring of chemical experiments remains challenging in transparent glassware, where weak phase boundaries and optical artifacts degrade conventional segmentation. We formulate laboratory phenomena as the time evolution of phase interfaces and introduce the Chemical Transparent Glasses dataset 2.0 (CTG 2.0), a vessel-aware benchmark with 3,668 images, 23 glassware categories, and five multiphase interface types for phase-interface instance segmentation. Building on YOLO11m-seg, we propose LGA-RCM-YOLO, which combines Local-Global Attention (LGA) for robust semantic representation and a Rectangular Self-Calibration Module (RCM) for boundary refinement of thin, elongated interfaces. On CTG 2.0, the proposed model achieves 84.4% AP@0.5 and 58.43% AP@0.5-0.95, improving over the YOLO11m baseline by 6.42 and 8.75 AP points, respectively, while maintaining near real-time inference (13.67 FPS, RTX 3060). An auxiliary color-attribute head further labels liquid instances as colored or colorless with 98.71% precision and 98.32% recall. Finally, we demonstrate continuous process monitoring in separatory-funnel phase separation and crystallization, showing that phase-interface instance segmentation can serve as a practical visual sensor for laboratory automation.

cs.CV↗

arXiv2026

Memorization capacity of deep ReLU neural networks characterized by width and depth

This paper studies the memorization capacity of deep neural networks with ReLU activation. Specifically, we investigate the minimal size of such networks to memorize any $N$ data points in the unit ball with pairwise separation distance $δ$ and discrete labels. Most prior studies characterize the memorization capacity by the number of parameters or neurons. We generalize these results by constructing neural networks, whose width $W$ and depth $L$ satisfy $W^2L^2= \mathcal{O}(N\log(δ^{-1}))$, that can memorize any $N$ data samples. We also prove that any such networks should also satisfy the lower bound $W^2L^2=Ω(N \log(δ^{-1}))$, which implies that our construction is optimal up to logarithmic factors when $δ^{-1}$ is polynomial in $N$. Hence, we explicitly characterize the trade-off between width and depth for the memorization capacity of deep neural networks in this regime.

cs.LG↗

arXiv2026

Artificial Intelligence for Detecting Fetal Orofacial Clefts and Advancing Medical Education

Orofacial clefts are among the most common congenital craniofacial abnormalities, yet accurate prenatal detection remains challenging due to the scarcity of experienced specialists and the relative rarity of the condition. Early and reliable diagnosis is essential to enable timely clinical intervention and reduce associated morbidity. Here we show that an artificial intelligence system, trained on over 45,139 ultrasound images from 9,215 fetuses across 22 hospitals, can diagnose fetal orofacial clefts with sensitivity and specificity exceeding 93% and 95% respectively, matching the performance of senior radiologists and substantially outperforming junior radiologists. When used as a medical copilot, the system raises junior radiologists' sensitivity by more than 6%. Beyond direct diagnostic assistance, the system also accelerates the development of clinical expertise. A pilot study involving 24 radiologists and trainees demonstrated that the model can improve the expertise development for rare conditions. This dual-purpose approach offers a scalable solution for improving both diagnostic accuracy and specialist training in settings where experienced radiologists are scarce.

cs.CV↗

arXiv2026

The Lie of the Average: How Class Incremental Learning Evaluation Deceives You?

Class Incremental Learning (CIL) requires models to continuously learn new classes without forgetting previously learned ones, while maintaining stable performance across all possible class sequences. In real-world settings, the order in which classes arrive is diverse and unpredictable, and model performance can vary substantially across different sequences. Yet mainstream evaluation protocols calculate mean and variance from only a small set of randomly sampled sequences. Our theoretical analysis and empirical results demonstrate that this sampling strategy fails to capture the full performance range, resulting in biased mean estimates and a severe underestimation of the true variance in the performance distribution. We therefore contend that a robust CIL evaluation protocol should accurately characterize and estimate the entire performance distribution. To this end, we introduce the concept of extreme sequences and provide theoretical justification for their crucial role in the reliable evaluation of CIL. Moreover, we observe a consistent positive correlation between inter-task similarity and model performance, a relation that can be leveraged to guide the search for extreme sequences. Building on these insights, we propose EDGE (Extreme case-based Distribution & Generalization Evaluation), an evaluation protocol that adaptively identifies and samples extreme class sequences using inter-task similarity, offering a closer approximation of the ground-truth performance distribution. Extensive experiments demonstrate that EDGE effectively captures performance extremes and yields more accurate estimates of distributional boundaries, providing actionable insights for model selection and robustness checking. Our code is available at https://github.com/AIGNLAI/EDGE.

cs.LG↗

arXiv2026

PromptStereo: Zero-Shot Stereo Matching via Structure and Motion Prompts

Modern stereo matching methods have leveraged monocular depth foundation models to achieve superior zero-shot generalization performance. However, most existing methods primarily focus on extracting robust features for cost volume construction or disparity initialization. At the same time, the iterative refinement stage, which is also crucial for zero-shot generalization, remains underexplored. Some methods treat monocular depth priors as guidance for iteration, but conventional GRU-based architectures struggle to exploit them due to the limited representation capacity. In this paper, we propose Prompt Recurrent Unit (PRU), a novel iterative refinement module based on the decoder of monocular depth foundation models. By integrating monocular structure and stereo motion cues as prompts into the decoder, PRU enriches the latent representations of monocular depth foundation models with absolute stereo-scale information while preserving their inherent monocular depth priors. Experiments demonstrate that our PromptStereo achieves state-of-the-art zero-shot generalization performance across multiple datasets, while maintaining comparable or faster inference speed. Our findings highlight prompt-guided iterative refinement as a promising direction for zero-shot stereo matching.

cs.CV↗

arXiv2026

Mean Effects on Critical Well-Posedness for Majda-Biello Systems on the Torus

This paper studies how the mean of the initial data $u_0$ affects the critical indices concerning local well-posedness for the following Majda-Biello systems: \[ \left\{\begin{aligned} & u_t + u_{xxx} + vv_x = 0 , \\ & v_t + αv_{xxx} + (uv)_x = 0 , \\ & (u,v) \mid_{t=0} = (u_0, v_0) \in H^s(\mathbb{T}) \times H^s(\mathbb{T}), \end{aligned}\right. \qquad x \in \mathbb{T}, \, t\in \mathbb{R}, \] where $\mathbb{T}$ refers to the periodic torus and the dispersion coefficient $α$ is restricted in $(0,4] \setminus \{1\}$ which corresponds to resonant cases. Previously, under the zero-mean assumption on $u_0$, Oh (Int. Math. Res. Not., (18):3516-3556, 2009) determined the critical indices $s^{*}(α)$ of the Sobolev regularity of the initial data for $C^3$ local well-posedness. In particular, Oh showed that \[ s^{*}(α) = \left\{ \begin{array}{lll} 1, & \text{for $α$ such that $\sqrt{12/α- 3} \in \mathbb{Q}$ }, \\ \frac12, & \text{for a.e. $α$ such that $\sqrt{12/α- 3} \notin \mathbb{Q}$ }. \end{array}\right. \] In this paper, by allowing the mean of $u_0$ to be non-zero, we find that the critical index $s^{*}(α)$ can be lowered from $1$ to $\frac12$ when $\sqrt{12/α- 3} \in \mathbb{Q}$. For other values of $α$, except in a set of zero measure, we also justify the critical index $s^{*}(α)$ to be $\frac12$ regardless of the mean of $u_0$. By subtracting the mean from $u_0$, the original Majda-Biello systems are slightly modified to contain first-order terms but with zero-mean initial data. The key ingredient in our proof is to introduce a refined Diophantine approximation theory to capture the essential resonance effect for the perturbed dispersive structure caused by these additional first-order terms.

math.AP↗

arXiv2026

nnLandmark: A Self-Configuring Method for 3D Medical Landmark Detection

Landmark detection is central to many medical applications, such as identifying critical structures for treatment planning or defining control points for biometric measurements. However, manual annotation is labor-intensive and requires expert anatomical knowledge. While deep learning shows promise in automating this task, fair evaluation and interpretation of methods in a broader context are hindered by limited public benchmarking, inconsistent baseline implementations, and non-standardized experimentation. To overcome these pitfalls, we present nnLandmark, a self-configuring framework for 3D landmark detection that combines tailored heatmap generation, loss design, inference logic, and a robust set of hyperparameters for heatmap regression, while reusing components from nnU-Net's underlying self-configuration and training engine. nnLandmark achieves state-of-the-art performance across five public and one private dataset, benchmarked against three recently published methods. Its out-of-the-box usability enables training strong landmark detection models on new datasets without expert knowledge or dataset-specific hyperparameter tuning. Beyond accuracy, nnLandmark provides both a strong, common baseline and a flexible, standardized environment for developing and evaluating new methodological contributions. It further streamlines evaluation across multiple datasets by offering data conversion utilities for current public benchmarks. Together, these properties position nnLandmark as a central tool for advancing 3D medical landmark detection through systematic, transparent benchmarking, enabling to genuinely measure methodological progress. The code is available on GitHub: https://github.com/MIC-DKFZ/nnLandmark

cs.CV↗

arXiv2026

Global well-posedness of the Majda-Biello system in the resonant case on the real line

We study the Cauchy problem for the following Majda-Biello system in the case $α=4$, where the resonance effect is the most significant, on the real line. \[ \left\{ \begin{array}{rcl} u_{t} + u_{xxx} & = & - v v_x, v_{t} + αv_{xxx} & = & - (uv)_{x}, (u,v)|_{t=0} & = & (u_0,v_0) \in H^{s}(\mathbb{R}) \times H^{s}(\mathbb{R}), \end{array} \right. \quad x \in \mathbb{R}, \, t \in \mathbb{R}. \] For Sobolev regularity $s\in[\frac34, 1)$, we establish global well-posedness by refining the I-method. Previously, the critical index for local well-posedness was known to be $\frac34$, while global well-posedness was only obtained for $s\geq 1$. Our global well-posedness result bridges the gap and matches the threshold in the local theory. The main novelty of our approach is to introduce a pair of distinct $I$-operators, tailored to the resonant structure of the Majda-Biello system with $α=4$. This dual-operator framework allows for pointwise control of the multipliers in the modified energies constructed via the multilinear correction technique. These modified energies are almost conserved and provide effective control over the Sobolev norm of the solution for all time. This new approach has potential applications to other coupled dispersive systems exhibiting strong resonant interactions.

math.AP↗

arXiv2026

Deep Generative Models in Condition and Structural Health Monitoring: Opportunities, Limitations and Future Outlook

Condition and structural health monitoring (CM/SHM) is a pivotal component of predictive maintenance (PdM) strategies across diverse industrial sectors, including mechanical rotating machinery, aircraft structures, wind turbines, and civil infrastructures. Conventional deep learning models, while effective for fault diagnosis and anomaly detection through automatic feature learning from sensor data, often struggle with operational variability, imbalanced or scarce fault datasets, and multimodal sensory data from complex systems. Deep generative models (DGMs) including deep autoregressive models, variational autoencoders, generative adversarial networks, diffusion-based models, and emerging large language models, offer transformative capabilities by synthesizing high-fidelity data samples, reconstructing latent system states, and modeling complex multimodal data streams. This review systematically examines state-of-the-art DGM applications in CM/SHM across the four main industrial systems mentioned above, emphasizing their roles in addressing key challenges: data generation, domain adaptation and generalization, multimodal data fusion, and downstream fault diagnosis and anomaly detection tasks, with rigorous comparison among signal processing, conventional machine learning or deep learning models, and DGMs. Lastly, we discuss current limitations of DGMs, including challenges of explainable and trustworthy models, computational inefficiencies for edge deployment, and the need for parameter-efficient fine-tuning strategies. Future research directions can focus on zero-shot and few-shot learning, robust multimodal data generation, hybrid architectures integrating DGMs with physics knowledge, and reinforcement learning with DGMs to enhance robustness and accuracy in industrial scenarios.

cs.CE↗

arXiv2026

Towards Self-Robust LLMs: Intrinsic Prompt Noise Resistance via CoIPO

Large language models (LLMs) have demonstrated remarkable and steadily improving performance across a wide range of tasks. However, LLM performance may be highly sensitive to prompt variations especially in scenarios with limited openness or strict output formatting requirements, indicating insufficient robustness. In real-world applications, user prompts provided to LLMs often contain imperfections, which may undermine the quality of the model's responses. To address this issue, previous work has primarily focused on preprocessing prompts, employing external tools or even LLMs to refine prompt formulations in advance. However, these approaches overlook the intrinsic robustness of LLMs, and their reliance on external components introduces additional computational overhead and uncertainty. In this work, we propose a Contrastive Learning-based Inverse Direct Preference Optimization (CoIPO) method that minimizes the discrepancy between the label-aligned logits produced by the model under a clean prompt and its noisy counterpart, and conduct a detailed analysis using mutual information theory. We augment the FLAN dataset by constructing paired prompts, each consisting of a clean prompt and its corresponding noisy version for training. Additionally, to evaluate the effectiveness, we develop NoisyPromptBench, a benchmark enhanced and derived from the existing PromptBench. Experimental results conducted on NoisyPromptBench demonstrate that our proposed method achieves a significant improvement in average accuracy over the current state-of-the-art approaches. The source code of CoIPO, pair-wise FLAN datasets, and NoisyPromptBench have already been released on https://github.com/vegetable-yx/CoIPO.

cs.CL↗

arXiv2026

Semantic Search At LinkedIn

Semantic search with large language models (LLMs) enables retrieval by meaning rather than keyword overlap, but scaling it requires major inference efficiency advances. We present LinkedIn's LLM-based semantic search framework for AI Job Search and AI People Search, combining an LLM relevance judge, embedding-based retrieval, and a compact Small Language Model trained via multi-teacher distillation to jointly optimize relevance and engagement. A prefill-oriented inference architecture co-designed with model pruning, context compression, and text-embedding hybrid interactions boosts ranking throughput by over 75x under a fixed latency constraint while preserving near-teacher-level NDCG, enabling one of the first production LLM-based ranking systems with efficiency comparable to traditional approaches and delivering significant gains in quality and user engagement.

cs.IR↗

arXiv2026

Beyond Classical Attention: Quantum Attention for Scalable Computation

As large language models (LLMs) demonstrate outstanding performance across various tasks, attention-driven models have profoundly transformed the field of machine learning. Since attention computations account for the primary computational overhead in both model inference and training, efficiently computing attention matrices has become one of the core challenges in accelerating large language models. It is well-known that quantum machines possess computational advantages over classical machines, and the role of quantum computing in LLMs remains largely unexplored. In this work, we focus on leveraging the Grover search algorithm to efficiently compute a sparse attention matrix. Through comparisons with classical algorithms, we demonstrate that our method achieves quantum acceleration in polynomial time. Additionally, we observe that the generated quantum attention matrices naturally exhibit low-rank structures, providing further theoretical support for efficient modeling. Moreover, within the specific context of attention matrix computation, we conduct a systematic and detailed analysis of the error and time complexity of the proposed algorithm.

quant-ph↗