Browsing by Author "Xiao, Wen"

Now showing 1 - 2 of 2
  • Thumbnail Image
    Item
    A new species of Notacanthella Jacobus & McCafferty, 2008 (Ephemeroptera, Ephemerellidae) from Yunnan, China
    (Pensoft, 2022) Li, Xian-Fu; Sun, Ye-Kang; Liu, Zi-Ye; Jacobus, Luke M.; Xiao, Wen; IUPUC Division of Science
    Notacanthella jinwu Li & Jacobus, sp. nov. is described based on egg, nymph, and winged stages from Dali Bai Autonomous Prefecture, Yunnan Province, China. The nymph of the new species is closely related to N. commodema (Allen, 1971), whose nymphs share a similar tuberculation of head, pronotum, and mesonotum. However, the nymph of our new species can be distinguished based on the structures of male sternum IX and abdominal tergal tubercles. In addition, the new species is distributed in subtropical high-altitude areas. The description of the male imago of the new species is the first certain one for the genus Notacanthella. Data associated with our new species allow for expanded discussion and diagnosis of Notacanthella and closely related genera. An identification key for nymphs of these groups is provided.
  • Thumbnail Image
    Item
    T3-Vis: a visual analytic framework for Training and fine-Tuning Transformers in NLP
    (ACL Anthology, 2021) Li, Raymond; Xiao, Wen; Wang, Lanjun; Jang, Hyeju; Carenini, Giuseppe; Computer Science, Luddy School of Informatics, Computing, and Engineering
    Transformers are the dominant architecture in NLP, but their training and fine-tuning is still very challenging. In this paper, we present the design and implementation of a visual analytic framework for assisting researchers in such process, by providing them with valuable insights about the model’s intrinsic properties and behaviours. Our framework offers an intuitive overview that allows the user to explore different facets of the model (e.g., hidden states, attention) through interactive visualization, and allows a suite of built-in algorithms that compute the importance of model components and different parts of the input sequence. Case studies and feedback from a user focus group indicate that the framework is useful, and suggest several improvements. Our framework is available at: https://github.com/raymondzmc/T3-Vis.