arXiv / html_feedback

Supports a student project developing a UI for feedback on arXiv articles rendered as html.
MIT License
17 stars 3 forks source link

Inconsistent font colors #1044

Open joyliu9562 opened 5 months ago

joyliu9562 commented 5 months ago

Description

On my windows laptop, dark mode, font in some paragraphs is light while some is dark. Even with in one paragraph, some sentences is dark and some is light. BTW, I have to say that great equation display!.

(Optional:) Please add any files, screenshots, or other information here.

No response

(Required) What is this issue most closely related to? Select one.

Choose One

Internal issue ID

5f54d43b-43c6-4613-9f49-108f7579a114

Paper URL

https://arxiv.org/html/2403.12448v1

Browser

Chrome/123.0.0.0

Device Type

Desktop

html-feedback-bot[bot] commented 5 months ago

Location in document: undefined

Selected HTML: ier with weights Bβˆˆβ„kΓ—r𝐡superscriptβ„π‘˜π‘ŸB\in{\mathbb{R}}^{k\times r}italic_B ∈ blackboard_R start_POSTSUPERSCRIPT italic_k Γ— italic_r end_POSTSUPERSCRIPT (rπ‘Ÿritalicr denotes the number of classes) as <math alttext="{g}{f,B}" class="ltxMath" display="inline" id="S4.SS1.p3.3.3.m3.2">gf,Bsubscript𝑔𝑓𝐡{g}{f,B}italic_g start_POSTSUBSCRIPT italic_f , italic_B end_POSTSUBSCRIPT on top of pretrained features f⁒(x)βˆˆβ„k𝑓π‘₯superscriptβ„π‘˜f(x)\in{\mathbb{R}}^{k}italic_f ( italic_x ) ∈ blackboard_R start_POSTSUPERSCRIPT italic_k end_POSTSUPERSCRIPT to predict the labels yβˆˆπ’΄π‘¦π’΄y\in{\mathcal{Y}}italic_y ∈ caligraphicY of augmented data. Then we define a majority voting classifier <math alttext="\bar{g}{f,B}(\bar{x}):={\operatorname{arg\,max}{{i\in[r]}}}\operatorname{Pr% }{x\sim\mathcal{A}(\cdot\mid\bar{x})}\left(g_{f,B}(x)=i\right)" class="ltx_math_unparsed" display="inline" id="S4.SS1.p3.6.6.m6.9">gΒ―f,B⁒(xΒ―):=arg⁒maxi∈[r]⁑PrxβˆΌπ’œ(β‹…βˆ£xΒ―)⁑(gf,B⁒(x)=i)\bar{g}_{f,B}(\bar{x}):={\operatorname{arg\,max}{{i\in[r]}}}\operatorname{Pr% }{x\sim\mathcal{A}(\cdot\mid\bar{x})}\left(g_{f,B}(x)=i\right)overΒ― start_ARG italic_g end_ARG start_POSTSUBSCRIPT italic_f , italic_B end_POSTSUBSCRIPT ( overΒ― start_ARG italic_x end_ARG ) := start_OPERATOR roman_arg roman_max end_OPERATOR start_POSTSUBSCRIPT italic_i ∈ [ italic_r ] end_POSTSUBSCRIPT roman_Pr start_POSTSUBSCRIPT italic_x ∼ caligraphic_A ( β‹… ∣ overΒ― start_ARG italic_x end_ARG ) end_POSTSUBSCRIPT ( italic_g start_POSTSUBSCRIPT italic_f , italic_B end_POSTSUBSCRIPT ( italic_x ) = italic_i ) to predict real data. A smaller classification error, denoted as ℰ⁒(f,B)ℰ𝑓𝐡{\mathcal{E}}(f,B)caligraphic_E ( italic_f , italic_B ), indicates be

github-actions[bot] commented 5 months ago

Hello @joyliu9562, thanks for the issue report! We are reviewing your report and will address it as soon as possible.