Yijia Zhang

Dr. Yijia Zhang


Research Focus & Metrics

Academic Career Length

2Years (2025-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
24 May 2025 in Mathematics
Verified
Journal Article
Open access
PDF file
YZYijia Zhang
Yijia Zhang
超刘超凡 刘
超凡 刘
YZYuan Zhu
Yuan Zhu
WCWanyu Chen
Wanyu Chen
Extracting causal relations from natural language texts is crucial for uncovering causality, and most existing causal relation extraction models are single-task learning-based models, which can not comprehensively address attributes such as part-of-speech tagging and chunk analysis. However, the characteristics of words with multi-domains are more relevant for causal relation extraction, due to words such as adjectives, linking verbs, etc., bringing more noise data limiting the effectiveness of the single-task-based learning methods. Furthermore, causalities from diverse domains also raise a challenge, as existing models tend to falter in multiple domains compared to a single one. In light of this, we propose a multi-task sequence tagging model, MPC−CE, which utilizes more information about causality and relevant tasks to improve causal relation extraction in noised data. By modeling auxiliary tasks, MPC−CE promotes a hierarchical understanding of linguistic structure and semantic roles, filtering noise and isolating salient entities. Furthermore, the sparse sharing paradigm extracts only the most broadly beneficial parameters by pruning redundant ones during training, enhancing model generalization. The empirical results on two datasets show 2.19% and 3.12% F1 improvement, respectively, compared to baselines, demonstrating that our proposed model can effectively enhance causal relation extraction with semantic features across multiple syntactic tasks, offering the representational power to overcome pervasive noise and cross-domain issues.

Verified
Journal Article
Open access
PDF file
Yanying MaoYanying Mao
Yanying Mao
YZYijia Zhang
Yijia Zhang
TSTaihua Shao
Taihua Shao
HCHonghui Chen
Honghui Chen
Empathetic response generation stands as a pivotal endeavor in the development of human-like dialogue systems. An effective approach in previous research is integrating external knowledge to generate empathetic responses. However, existing approaches only focus on identifying a user’s current emotional state, and they overlook the user’s emotional transition during context, and fail to propel the sustainability of the dialogue. To tackle the aforementioned issues, we propose an empathetic response generation model based on an emotional transition prompt and dual-semantic contrastive learning (EPDC). Specifically, we first compute the transition in users’ sentiment polarity during the conversation and incorporate it into the conversation embedding as sentiment prompts. Then, we generate two distinct fine-grained contextual representations and treat them as positive examples for contrastive learning, respectively, aiming at extracting high-order semantic information to guide the subsequent turn of dialogue. Finally, we also leverage commonsense knowledge to enhance the contextual representations, and the empathetic responses are generated by decoding the combination of semantic and emotional states. Notably, our work represents the pioneering application of emotional prompts and contrastive learning to augment the sustainability of empathetic dialogue. Extensive experiments conducted on the benchmark dataset EMPATHETICDIALOGUES demonstrate that EPDC outperforms the baselines in both automatic evaluations and human evaluations.

9 May 2026 in Electronics
Verified
Journal Article
Open access
PDF file
ZCZhiwen Chen
Zhiwen Chen
YZYijia Zhang
Yijia Zhang
MLMiao Liu
Miao Liu
YPYue Peng
Yue Peng
Flat label supervision often constrains multi-label image classification, as it struggles to fully capture inherent label dependencies. It provides limited guidance to the hierarchical features that naturally emerge in Vision Transformers. To address this structural misalignment, we propose Dynamic Topic-based Hierarchical Prompt Learning (DyT-HPL). Instead of relying on predefined and fixed label graphs, DyT-HPL utilizes offline hierarchical clustering to construct multi-granularity semantic priors, from which hierarchical prompts are dynamically retrieved. A frozen visual query branch generates stable semantic queries, which are then used to retrieve discrete prompts from constructed coarse-mid-fine prompt pools. These hierarchical prompts are adaptively injected into different network depths, ensuring that semantic guidance with different abstraction levels is introduced at the most suitable architectural stages. To maintain stable routing and prevent prompt mode collapse, we jointly optimize the architecture with asymmetric classification, surrogate matching, and intra-pool diversity losses. This tripartite design promotes a diverse prompt space and isolates routing updates from final predictions. Comprehensive experiments on MS-COCO, NUS-WIDE, and Corel5k demonstrate that DyT-HPL achieves consistent and favorable performance across diverse settings, highlighting the value of hierarchical semantic guidance with different abstraction levels.

20 July 2026 in Electronics
Verified
Journal Article
Open access
PDF file
MLMiao Liu
Miao Liu
JWJiaying Wu
Jiaying Wu
MKMingxing Ke
Mingxing Ke
YZYijia Zhang
Yijia Zhang
This paper develops a stability-certified distributed residual neural network (ResNN) controller for leaderless consensus of continuous-time first-order multi-agent systems over undirected connected graphs. Unlikeconventional linear protocols and post-training verification methods, the proposed framework combines nonlinear transient shaping with certificate feasibility throughout learning. Each agent uses relative state differences with neighboring agents and applies a shared ResNN policy composed of a linear diffusive backbone and a bias-free neural residual branch. A computable residual-gain bound yields a sufficient condition for global exponential consensus, with the certified rate determined by the backbone–residual gain margin and the algebraic connectivity of the graph. Spectral-norm projection and gain reparameterization preserve this condition after every training update. A complementary forward-Euler certificate provides an explicit step-size condition for exponentially convergent digital implementation. The controller is trained through differentiable closed-loop simulation using an objective that balances disagreement and control effort. Ournumerical results verify the certified decay bound, demonstrate competitive transient trade-offs against three baselines, and evaluate direct deployment across different graph topologies and network sizes.