Výsledky vyhledávání - "ZHANG Chunyun"

Report

Dynamic Prompt Allocation and Tuning for Continual Test-Time Adaptation

Autor: Cui, Chaoran, Zhen, Yongrui, Gong, Shuai, Zhang, Chunyun, Liu, Hui, Yin, Yilong

Continual test-time adaptation (CTTA) has recently emerged to adapt a pre-trained source model to continuously evolving target distributions, which accommodates the dynamic nature of real-world environments. To mitigate the risk of catastrophic forge

Externí odkaz: http://arxiv.org/abs/2412.09308

Zobrazit plný text záznamu

Report

Federated Domain Generalization via Prompt Learning and Aggregation

Autor: Gong, Shuai, Cui, Chaoran, Zhang, Chunyun, Wang, Wenna, Nie, Xiushan, Zhu, Lei

Federated domain generalization (FedDG) aims to improve the global model generalization in unseen domains by addressing data heterogeneity under privacy-preserving constraints. A common strategy in existing FedDG studies involves sharing domain-speci

Externí odkaz: http://arxiv.org/abs/2411.10063

Zobrazit plný text záznamu

Report

Do We Fully Understand Students' Knowledge States? Identifying and Mitigating Answer Bias in Knowledge Tracing

Autor: Cui, Chaoran, Ma, Hebo, Zhang, Chen, Zhang, Chunyun, Yao, Yumo, Chen, Meng, Ma, Yuling

Knowledge tracing (KT) aims to monitor students' evolving knowledge states through their learning interactions with concept-related questions, and can be indirectly evaluated by predicting how students will perform on future questions. In this paper,

Externí odkaz: http://arxiv.org/abs/2308.07779

Zobrazit plný text záznamu

Report

DGEKT: A Dual Graph Ensemble Learning Method for Knowledge Tracing

Autor: Cui, Chaoran, Yao, Yumo, Zhang, Chunyun, Ma, Hebo, Ma, Yuling, Ren, Zhaochun, Zhang, Chen, Ko, James

Knowledge tracing aims to trace students' evolving knowledge states by predicting their future performance on concept-related exercises. Recently, some graph-based models have been developed to incorporate the relationships between exercises to impro

Externí odkaz: http://arxiv.org/abs/2211.12881

Zobrazit plný text záznamu

Report

A Robust Contrastive Alignment Method For Multi-Domain Text Classification

Autor: Li, Xuefeng, Lei, Hao, Wang, Liwen, Dong, Guanting, Zhao, Jinzheng, Liu, Jiachi, Xu, Weiran, Zhang, Chunyun

Multi-domain text classification can automatically classify texts in various scenarios. Due to the diversity of human languages, texts with the same label in different domains may differ greatly, which brings challenges to the multi-domain text class

Externí odkaz: http://arxiv.org/abs/2204.12125

Zobrazit plný text záznamu