Duzhen Zhang
YOU?
Author Swipe
View article: MedKGent: A Large Language Model Agent Framework for Constructing Temporally Evolving Medical Knowledge Graph
MedKGent: A Large Language Model Agent Framework for Constructing Temporally Evolving Medical Knowledge Graph Open
The rapid expansion of medical literature presents growing challenges for structuring and integrating domain knowledge at scale. Knowledge Graphs (KGs) offer a promising solution by enabling efficient retrieval, automated reasoning, and kn…
View article: Federated Incremental Named Entity Recognition
Federated Incremental Named Entity Recognition Open
Federated Named Entity Recognition (FNER) boosts model training within each local client by aggregating the model updates of decentralized local clients, without sharing their private data. However, existing FNER methods assume fixed entit…
View article: How to Continually Adapt Text-to-Image Diffusion Models for Flexible Customization?
How to Continually Adapt Text-to-Image Diffusion Models for Flexible Customization? Open
Custom diffusion models (CDMs) have attracted widespread attention due to their astonishing generative ability for personalized concepts. However, most existing CDMs unreasonably assume that personalized concepts are fixed and cannot chang…
View article: MM-LLMs: Recent Advances in MultiModal Large Language Models
MM-LLMs: Recent Advances in MultiModal Large Language Models Open
In the past year, MultiModal Large Language Models (MM-LLMs) have undergone substantial advancements, augmenting off-the-shelf LLMs to support MM inputs or outputs via cost-effective training strategies. The resulting models not only prese…
View article: Continual Named Entity Recognition without Catastrophic Forgetting
Continual Named Entity Recognition without Catastrophic Forgetting Open
Continual Named Entity Recognition (CNER) is a burgeoning area, which involves updating an existing model by incorporating new entity types sequentially. Nevertheless, continual learning approaches are often severely afflicted by catastrop…
View article: Task Relation Distillation and Prototypical Pseudo Label for Incremental Named Entity Recognition
Task Relation Distillation and Prototypical Pseudo Label for Incremental Named Entity Recognition Open
Incremental Named Entity Recognition (INER) involves the sequential learning of new entity types without accessing the training data of previously learned types. However, INER faces the challenge of catastrophic forgetting specific for inc…
View article: Federated Incremental Semantic Segmentation
Federated Incremental Semantic Segmentation Open
Federated learning-based semantic segmentation (FSS) has drawn widespread attention via decentralized training on local clients. However, most FSS models assume categories are fixed in advance, thus heavily undergoing forgetting on old cat…
View article: Crucial Semantic Classifier-based Adversarial Learning for Unsupervised Domain Adaptation
Crucial Semantic Classifier-based Adversarial Learning for Unsupervised Domain Adaptation Open
Unsupervised Domain Adaptation (UDA), which aims to explore the transferrable features from a well-labeled source domain to a related unlabeled target domain, has been widely progressed. Nevertheless, as one of the mainstream, existing adv…
View article: VLP: A Survey on Vision-language Pre-training
VLP: A Survey on Vision-language Pre-training Open
View article: Continual Named Entity Recognition without Catastrophic Forgetting
Continual Named Entity Recognition without Catastrophic Forgetting Open
Continual Named Entity Recognition (CNER) is a burgeoning area, which involves updating an existing model by incorporating new entity types sequentially. Nevertheless, continual learning approaches are often severely afflicted by catastrop…
View article: DualGATs: Dual Graph Attention Networks for Emotion Recognition in Conversations
DualGATs: Dual Graph Attention Networks for Emotion Recognition in Conversations Open
Capturing complex contextual dependencies plays a vital role in Emotion Recognition in Conversations (ERC). Previous studies have predominantly focused on speaker-aware context modeling, overlooking the discourse structure of the conversat…
View article: Tuning Synaptic Connections instead of Weights by Genetic Algorithm in Spiking Policy Network
Tuning Synaptic Connections instead of Weights by Genetic Algorithm in Spiking Policy Network Open
Learning from interaction is the primary way that biological agents acquire knowledge about their environment and themselves. Modern deep reinforcement learning (DRL) explores a computational approach to learning from interaction and has m…
View article: HiVLP: Hierarchical Vision-Language Pre-Training for Fast Image-Text Retrieval
HiVLP: Hierarchical Vision-Language Pre-Training for Fast Image-Text Retrieval Open
In the past few years, the emergence of vision-language pre-training (VLP) has brought cross-modal retrieval to a new era. However, due to the latency and computation demand, it is commonly challenging to apply VLP in a real-time online re…
View article: TSAM: A Two-Stream Attention Model for Causal Emotion Entailment
TSAM: A Two-Stream Attention Model for Causal Emotion Entailment Open
Causal Emotion Entailment (CEE) aims to discover the potential causes behind an emotion in a conversational utterance. Previous works formalize CEE as independent utterance pair classification problems, with emotion and speaker information…
View article: A Visual Attention Based Object Detection Model beyond Top-Down and Bottom-up Mechanism
A Visual Attention Based Object Detection Model beyond Top-Down and Bottom-up Mechanism Open
\nTraditional saliency-based attention theory supposed that bottom-up and top-down factors combine to direct attentional behavior. This dichotomy fails to explain a growing number of cases in which neither bottom-up nor top-down can accoun…