Zobrazeno 1 - 10
of 1 178
pro vyhledávání: '"Liu, Yiyang"'
In the facial expression recognition task, researchers always get low accuracy of expression classification due to a small amount of training samples. In order to solve this kind of problem, we proposes a new data augmentation method named MixCut. In
Externí odkaz:
http://arxiv.org/abs/2405.10489
Autor:
Ren, Qingyang, Jiang, Zilin, Cao, Jinghan, Li, Sijia, Li, Chiqu, Liu, Yiyang, Huo, Shuning, He, Tiange, Chen, Yuan
This survey explores the fairness of large language models (LLMs) in e-commerce, examining their progress, applications, and the challenges they face. LLMs have become pivotal in the e-commerce domain, offering innovative solutions and enhancing cust
Externí odkaz:
http://arxiv.org/abs/2405.13025
Autor:
Liu, Kaibo, Liu, Yiyang, Chen, Zhenpeng, Zhang, Jie M., Han, Yudong, Ma, Yun, Li, Ge, Huang, Gang
Conventional automated test generation tools struggle to generate test oracles and tricky bug-revealing test inputs. Large Language Models (LLMs) can be prompted to produce test inputs and oracles for a program directly, but the precision of the test
Externí odkaz:
http://arxiv.org/abs/2404.10304
Autor:
Zhang, Qiangbo, Yu, Zeqing, Wang, Mengguang, Liu, Yiyang, Zhang, Changwei, Wang, Chang, Zheng, Zhenrong
Single metalenses are limited by their physical constraints, precluding themselves from achieving high numerical aperture across a wide visible spectral band in large-aperture applications. A hybrid system that integrates a metalens with a refractive
Externí odkaz:
http://arxiv.org/abs/2404.03173
The inverse wave scattering problem seeks to estimate a heterogeneous, inaccessible medium, modeled by unknown variable coefficients in wave equations, from transient recordings of waves generated by probing signals. It is a widely studied inverse pr
Externí odkaz:
http://arxiv.org/abs/2403.03844
Knowledge Distillation (KD) transfers the knowledge from a high-capacity teacher model to promote a smaller student model. Existing efforts guide the distillation by matching their prediction logits, feature embedding, etc., while leaving how to effi
Externí odkaz:
http://arxiv.org/abs/2211.17059
Many recent named entity recognition (NER) studies criticize flat NER for its non-overlapping assumption, and switch to investigating nested NER. However, existing nested NER models heavily rely on training data annotated with nested entities, while
Externí odkaz:
http://arxiv.org/abs/2211.00301
Span-based models are one of the most straightforward methods for named entity recognition (NER). Existing span-based NER systems shallowly aggregate the token representations to span representations. However, this typically results in significant in
Externí odkaz:
http://arxiv.org/abs/2210.04182
Autor:
Wang, Qin1 (AUTHOR) qinwang997@163.com, Liu, Yiyang2 (AUTHOR) 1020210185@glut.edu.cn, Zhang, Baolin2 (AUTHOR) baolinzhang@ymail.com, Dong, Jianghui1 (AUTHOR) djh1028@163.com, Wang, Liping1 (AUTHOR) djh1028@163.com
Publikováno v:
Polymers (20734360). Jun2024, Vol. 16 Issue 11, p1457. 22p.
Publikováno v:
In Applied Acoustics 5 July 2024 223