arXiv subjects

Liying Wang

Publications and source records attributed to Liying Wang.

14 recordsLinked to original sources

arXiv2026

Initial Risk Probing and Feasibility Testing of Glow: a Generative AI-Powered Dialectical Behavior Therapy Skills Coach for Substance Use Recovery and HIV Prevention

Background: HIV and substance use represent interacting epidemics with shared psychological drivers - impulsivity and maladaptive coping. Dialectical behavior therapy (DBT) targets these mechanisms but faces scalability challenges. Generative artificial intelligence (GenAI) offers potential for delivering personalized DBT coaching at scale, yet rapid development has outpaced safety infrastructure. Methods: We developed Glow, a GenAI-powered DBT skills coach delivering chain and solution analysis for individuals at risk for HIV and substance use. In partnership with a Los Angeles community health organization, we conducted usability testing with clinical staff (n=6) and individuals with lived experience (n=28). Using the Helpful, Honest, and Harmless (HHH) framework, we employed user-driven adversarial testing wherein participants identified target behaviors and generated contextually realistic risk probes. We evaluated safety performance across 37 risk probe interactions. Results: Glow appropriately handled 73% of risk probes, but performance varied by agent. The solution analysis agent demonstrated 90% appropriate handling versus 44% for the chain analysis agent. Safety failures clustered around encouraging substance use and normalizing harmful behaviors. The chain analysis agent fell into an "empathy trap," providing validation that reinforced maladaptive beliefs. Additionally, 27 instances of DBT skill misinformation were identified. Conclusions: This study provides the first systematic safety evaluation of GenAI-delivered DBT coaching for HIV and substance use risk reduction. Findings reveal vulnerabilities requiring mitigation before clinical trials. The HHH framework and user-driven adversarial testing offer replicable methods for evaluating GenAI mental health interventions.

cs.AI↗

arXiv2025

Query Drift Compensation: Enabling Compatibility in Continual Learning of Retrieval Embedding Models

Text embedding models enable semantic search, powering several NLP applications like Retrieval Augmented Generation by efficient information retrieval (IR). However, text embedding models are commonly studied in scenarios where the training data is static, thus limiting its applications to dynamic scenarios where new training data emerges over time. IR methods generally encode a huge corpus of documents to low-dimensional embeddings and store them in a database index. During retrieval, a semantic search over the corpus is performed and the document whose embedding is most similar to the query embedding is returned. When updating an embedding model with new training data, using the already indexed corpus is suboptimal due to the non-compatibility issue, since the model which was used to obtain the embeddings of the corpus has changed. While re-indexing of old corpus documents using the updated model enables compatibility, it requires much higher computation and time. Thus, it is critical to study how the already indexed corpus can still be effectively used without the need of re-indexing. In this work, we establish a continual learning benchmark with large-scale datasets and continually train dense retrieval embedding models on query-document pairs from new datasets in each task and observe forgetting on old tasks due to significant drift of embeddings. We employ embedding distillation on both query and document embeddings to maintain stability and propose a novel query drift compensation method during retrieval to project new model query embeddings to the old embedding space. This enables compatibility with previously indexed corpus embeddings extracted using the old model and thus reduces the forgetting. We show that the proposed method significantly improves performance without any re-indexing. Code is available at https://github.com/dipamgoswami/QDC.

cs.IR↗

arXiv2025

MAFS: Masked Autoencoder for Infrared-Visible Image Fusion and Semantic Segmentation

Infrared-visible image fusion methods aim at generating fused images with good visual quality and also facilitate the performance of high-level tasks. Indeed, existing semantic-driven methods have considered semantic information injection for downstream applications. However, none of them investigates the potential for reciprocal promotion between pixel-wise image fusion and cross-modal feature fusion perception tasks from a macroscopic task-level perspective. To address this limitation, we propose a unified network for image fusion and semantic segmentation. MAFS is a parallel structure, containing a fusion sub-network and a segmentation sub-network. On the one hand, We devise a heterogeneous feature fusion strategy to enhance semantic-aware capabilities for image fusion. On the other hand, by cascading the fusion sub-network and a segmentation backbone, segmentation-related knowledge is transferred to promote feature-level fusion-based segmentation. Within the framework, we design a novel multi-stage Transformer decoder to aggregate fine-grained multi-scale fused features efficiently. Additionally, a dynamic factor based on the max-min fairness allocation principle is introduced to generate adaptive weights of two tasks and guarantee smooth training in a multi-task manner. Extensive experiments demonstrate that our approach achieves competitive results compared with state-of-the-art methods. The code is available at https://github.com/Abraham-Einstein/MAFS/.

cs.CV↗

arXiv2025

Large Language Model-Powered Conversational Agent Delivering Problem-Solving Therapy (PST) for Family Caregivers: Enhancing Empathy and Therapeutic Alliance Using In-Context Learning

Family caregivers often face substantial mental health challenges due to their multifaceted roles and limited resources. This study explored the potential of a large language model (LLM)-powered conversational agent to deliver evidence-based mental health support for caregivers, specifically Problem-Solving Therapy (PST) integrated with Motivational Interviewing (MI) and Behavioral Chain Analysis (BCA). A within-subject experiment was conducted with 28 caregivers interacting with four LLM configurations to evaluate empathy and therapeutic alliance. The best-performing models incorporated Few-Shot and Retrieval-Augmented Generation (RAG) prompting techniques, alongside clinician-curated examples. The models showed improved contextual understanding and personalized support, as reflected by qualitative responses and quantitative ratings on perceived empathy and therapeutic alliances. Participants valued the model's ability to validate emotions, explore unexpressed feelings, and provide actionable strategies. However, balancing thorough assessment with efficient advice delivery remains a challenge. This work highlights the potential of LLMs in delivering empathetic and tailored support for family caregivers.

cs.AI↗

arXiv2025

A Semantic-Aware and Multi-Guided Network for Infrared-Visible Image Fusion

Multi-modality image fusion aims at fusing modality-specific (complementarity) and modality-shared (correlation) information from multiple source images. To tackle the problem of the neglect of inter-feature relationships, high-frequency information loss, and the limited attention to downstream tasks, this paper focuses on how to model correlation-driven decomposing features and reason high-level graph representation by efficiently extracting complementary information and aggregating multi-guided features. We propose a three-branch encoder-decoder architecture along with corresponding fusion layers as the fusion strategy. Firstly, shallow features from individual modalities are extracted by a depthwise convolution layer combined with the transformer block. In the three parallel branches of the encoder, Cross Attention and Invertible Block (CAI) extracts local features and preserves high-frequency texture details. Base Feature Extraction Module (BFE) captures long-range dependencies and enhances modality-shared information. Graph Reasoning Module (GR) is introduced to reason high-level cross-modality relations and simultaneously extract low-level detail features as CAI's modality-specific complementary information. Experiments demonstrate the competitive results compared with state-of-the-art methods in visible/infrared image fusion and medical image fusion tasks. Moreover, the proposed algorithm surpasses the state-of-the-art methods in terms of subsequent tasks, averagely scoring 8.27% mAP@0.5 higher in object detection and 5.85% mIoU higher in semantic segmentation. The code is avaliable at https://github.com/Abraham-Einstein/SMFNet/.

cs.CV↗

arXiv2024

Deep Functional Factor Models: Forecasting High-Dimensional Functional Time Series via Bayesian Nonparametric Factorization

This paper introduces the Deep Functional Factor Model (DF2M), a Bayesian nonparametric model designed for analysis of high-dimensional functional time series. DF2M is built upon the Indian Buffet Process and the multi-task Gaussian Process, incorporating a deep kernel function that captures non-Markovian and nonlinear temporal dynamics. Unlike many black-box deep learning models, DF2M offers an explainable approach to utilizing neural networks by constructing a factor model and integrating deep neural networks within the kernel function. Additionally, we develop a computationally efficient variational inference algorithm to infer DF2M. Empirical results from four real-world datasets demonstrate that DF2M provides better explainability and superior predictive accuracy compared to conventional deep learning models for high-dimensional functional time series.

stat.ML↗

arXiv2024

Multi-perspective Feedback-attention Coupling Model for Continuous-time Dynamic Graphs

Recently, representation learning over graph networks has gained popularity, with various models showing promising results. Despite this, several challenges persist: 1) most methods are designed for static or discrete-time dynamic graphs; 2) existing continuous-time dynamic graph algorithms focus on a single evolving perspective; and 3) many continuous-time dynamic graph approaches necessitate numerous temporal neighbors to capture long-term dependencies. In response, this paper introduces the Multi-Perspective Feedback-Attention Coupling (MPFA) model. MPFA incorporates information from both evolving and raw perspectives, efficiently learning the interleaved dynamics of observed processes. The evolving perspective employs temporal self-attention to distinguish continuously evolving temporal neighbors for information aggregation. Through dynamic updates, this perspective can capture long-term dependencies using a small number of temporal neighbors. Meanwhile, the raw perspective utilizes a feedback attention module with growth characteristic coefficients to aggregate raw neighborhood information. Experimental results on a self-organizing dataset and seven public datasets validate the efficacy and competitiveness of our proposed model.

cs.LG↗

arXiv2024

Ideal spin-polarized Weyl-half-semimetal with a single pair of Weyl points in half-Heusler compounds XCrTe (X=K, Rb)

Realizing ideal Weyl semimetal state with a single pair of Weyl points has been a long-sought goal in the field of topological semimetals. Here, we reveal such a state in the Cr-based half-Heusler compounds XCrTe (X=K, Rb). We show that these materials have a half metal ground state, with Fermi level crossing only one spin channel. Importantly, the Fermi surface is clean, consisting of the minimal number (i.e., a single pair) of spin-polarized Weyl points, so the state represents an ideal Weyl half semimetal. We show that the locations of the two Weyl points and the associated Chern vector can be flexibly tuned by rotating the magnetization vector. The minimal surface Fermi arc pattern and its contribution to anomalous Hall transport are discussed. Our finding offers an ideal material platform for exploring magnetic Weyl fermions, which will also facilitate the interplay between Weyl physics and spintronics.

cond-mat.mes-hall↗

arXiv2023

EEGNN: Edge Enhanced Graph Neural Network with a Bayesian Nonparametric Graph Model

Training deep graph neural networks (GNNs) poses a challenging task, as the performance of GNNs may suffer from the number of hidden message-passing layers. The literature has focused on the proposals of {over-smoothing} and {under-reaching} to explain the performance deterioration of deep GNNs. In this paper, we propose a new explanation for such deteriorated performance phenomenon, {mis-simplification}, that is, mistakenly simplifying graphs by preventing self-loops and forcing edges to be unweighted. We show that such simplifying can reduce the potential of message-passing layers to capture the structural information of graphs. In view of this, we propose a new framework, edge enhanced graph neural network (EEGNN). EEGNN uses the structural information extracted from the proposed Dirichlet mixture Poisson graph model (DMPGM), a Bayesian nonparametric model for graphs, to improve the performance of various deep message-passing GNNs. We propose a Markov chain Monte Carlo inference framework for DMPGM. Experiments over different datasets show that our method achieves considerable performance increase compared to baselines.

stat.ML↗

arXiv2021

Intersonic Detachment Surface Waves in Elastomer Frictional Sliding

Elastomeric materials when sliding on clean and rough surfaces generate wrinkles at the interface due to tangential stress gradients. These interfacial folds travel along the bottom of elastomer as surface detachment waves to facilitate the apparent sliding motion of elastomer. At very low sliding speed compared to elastic surface waves, the process is dominated by surface adhesion and relaxation effects, and the phenomenon is historically referred to as Schallamach waves. We report in this letter the observation of fast-traveling intersonic detachment waves exceeding the Rayleigh and shear wave velocities of the soft material in contact. The spatio-temporal analysis revealed the accelerating nature of the detachment wave, and the scaling of wave speed with the elastic modului of the material suggests that this process is governed by elasticity and inertia. Multiple wave signatures on the plot were connected to different stages of surface wrinkles, as they exhibited distinctive slopes (from which velocities were derived) in the generation, propagation and rebound phases. We also characterized the frequencies of wrinkle generation in addition to the speeds and found a consistent scaling law of these two wave characteristics as the stiffness of elastomer increased. Physical implications of this new finding may further promote our understanding of elastomer noise generation mechanisms, as at macroscopic sliding velocity, the frequency of elastomer instability readily enters human audible ranges and interacts with other vibratory frequencies to cooperatively create harsh and detrimental noises in disc braking, wiper blade and shoe squeaking among many other elastomer applications.

cond-mat.soft↗

arXiv2021

Ideal fully spin-polarized type-II nodal line state in half-metals X2YZ4 (X=K, Cs, Rb, Y=Cr, Cu, Z=Cl, F)

Lorentz-violating type-II nodal lines exhibit attracting physical properties and have been hot discussed currently. However, their investigations have been mostly limited in nonmagnetic system because of lacking ideal spin-polarized candidates with clean type-II nodal line states. Here, for the first time, we report the family of X2YZ4 (X=K, Cs, Rb, Y=Cr, Cu, Z=Cl, F) compounds are such ideal candidate materials by using the member of K2CuF4 as an example. We show the material is a ferromagnetic half-metal with weak anisotropy, which host fully spin-polarized conducting electrons. In the conducting spin channel, the band crossing form a pair of type-II nodal lines, protected by mirror symmetry. These type-II nodal lines are different with former proposed examples because they have a 100% spin polarization. In addition, we also show the material can realize switchable topological states, which can be easily controlled by external magnetic field. It is noticed that, the material: i) is stable and can be synthesized in experiments; ii) has clear magnetic structure; and iii) manifests clean type-II nodal line state and clear drumhead surface states. Therefore, the proposed X2YZ4 compounds are expected to be an excellent platform to investigate the novel physical properties of both type-II nodal line states with complete

cond-mat.mtrl-sci↗

arXiv2020

Fully spin-polarized double-Weyl fermions with type-III dispersion in quasi-one dimensional materials X2RhF6 (X=K, Rb, Cs)

Double-Weyl fermions, as novel topological states of matter, have been mostly discussed in nonmagnetic materials. Here, based on density-functional theory and symmetry analysis, we propose the realization of fully spin-polarized double-Weyl fermions in a family ferromagnetic materials X2RhF6 (X= K, Rb, Cs). These materials have the half-metal ground states, where only the bands from the spin-down channel present near the Fermi energy. The spin-down bands form a pair of triply degenerate nodal points (TDNPs) if spin-orbit coupling (SOC) is not included. Under SOC, one TDNP splits into two double-Weyl points featuring quadratic dispersion along two momentum direction, and they are protected by the three-fold rotation (C3) symmetry. Unlike most double-Weyl semimetals, the Weyl points proposed here have the type-III dispersion with one of the crossing bands being saddle-shaped. An effective model is constructed, which describes well the nature of the Weyl points. These Weyl points are fully spin-polarized, and are characterized with double Fermi arcs on the surface spectrum. Breaking C3 symmetry by lattice strain could shift one double-Weyl point into a pair of type-II single-Weyl points. The X2RhF6 materials proposed here are excellent candidates to investigate the novel properties of type-III double-Weyl fermions in ferromagnetic system, as well as generate potential applications in spintronics.

cond-mat.mtrl-sci↗

arXiv2020

Two-dimensional Weyl nodal line semimetal in high Curie temperature d0 ferromagnet K2N monolayer

Nodal line semimetals in two-dimensional (2-D) materials have attracted intense attention currently. From fundamental physics and spintronic applications points of view, high Curie temperature ferromagnetic (FM) ones with nodal lines robust against spin-orbit coupling (SOC) are significantly in desirable. Here, we propose that FM K2N monolayer is such Weyl nodal line semimetal. We show that K2N monolayer is dynamically stable, and has a FM ground magnetic state with the out-of-plane [001] magnetization. It shows two nodal lines in the low-energy band structures. Both nodal lines are robust against SOC, under the protection of mirror symmetry. We construct an effective Hamiltonian, which can well characterize the nodal lines in the system. Remarkably, the nodal line semimetal proposed here is distinct from the previously studied ones in that K2N monolayer is 2-D d0-type ferromagnet with the magnetism arising from the partially filled N-p orbitals, which can bring special advantages in spintronic applications. Besides, the Curie temperature in K2N monolayer is estimated to be 942K, being significantly higher than previous FM nodal lines materials. We also find that, specific tensile strains can transform the nodal line from type-I to a type-II one, making its nodal line characteristics even more interesting.

cond-mat.mtrl-sci↗

arXiv2018

Hereditary effects of exponentially damped oscillators with past histories

Hereditary effects of exponentially damped oscillators with past histories are considered in this paper. Nonviscously damped oscillators involve hereditary damping forces which depend on time-histories of vibrating motions via convolution integrals over exponentially decaying functions. As a result, this kind of oscillators are said to have memory. In this work, initialization for nonviscously damped oscillators is firstly proposed. Unlike the classical viscously damped ones, information of the past history of response velocity is necessary to fully determine the dynamic behaviors of nonviscously damped oscillators. Then, initialization response of exponentially damped oscillators is obtained to characterize the hereditary effects on the dynamic response. At last, stability of initialization response is proved and the hereditary effects are shown to gradually recede with increasing of time.

math.DS↗