Super honored and excited that my work on the role of tech worker positionality in shaping identity concepts in computer vision was awarded a best paper 🤯
🎉 Congratulations to all authors honored with Best Paper Awards or Honorable Mentions at
#CHI2024
! We can't wait to celebrate with you in Hawaii soon! 🏄♂️🌴 Check out the award recipients here:
@sigchi
@ACMDL
We're excited to announce the 2021 Microsoft Research PhD Fellowship recipients! For over two decades, our PhD Fellowship has supported nearly 300 outstanding students pursuing exceptional research.
Learn more about our 10 new fellows:
SUPER excited to announce: "Auto-essentialization: Gender in automated facial analysis as extended colonial project" by myself and my amazing collaborators
@Madeleine_Pape
and
@alexhanna
where we critically analyze gender in facial analysis
Excited to finally announce: I have accepted a job with
@SonyAI_global
as a Research Scientist on their AI Ethics team! I have the amazing opportunity to be the first ever HCI RS on the team, working on topics like: ethical data collection, computer vision, and genAI.
I can post now! Our paper "How We've Taught Algorithms to See Identity" was awarded both a Best Paper Award and a Diversity & Inclusion Recognition award! Thank you to my amazing co-authors
@KandreaWade
@caitie_lustig
@whatknows
If you work on equity for marginalized groups in computing, you will inevitably experience pushback, indignance, and discrediting. This will sting even more if you identify as part of those groups. But it does not mean your work is not high quality, rigorous, or necessary.
Given all we know about the use of Uighur classification in China, I am extremely concerned to learn there is an IEEE proceedings paper with an associated database focused on classifying Tibetan, Uighur and Zhuang ethnic groups.
Preprint announcement: "Do Datasets Have Politics? Disciplinary Values in Computer Vision Dataset Development" w/ myself,
@cephaloponderer
,
@alexhanna
to be published in CSCW 2021
Being critical of a technology (e.g. facial analysis) is not about "burn it all down" & "criticism for criticism's sake." It is about building a better, more equitable world by calling out & dismantling the harm of said technology. Even if that means dismantling that technology.
Okay real talk. I've been very cognizant of the lack of cis voices saying anything about: (1) the loss of trans healthcare rights in the US and (2) the recent brutalization and killings of Black trans women. The silence is LOUD.
.
@morganklauss
.
@cephaloponderer
& .
@alexhanna
examined the underlying values of computer vision dataset practices in "Do Datasets Have Politics?" - contrasting them with otherwise silenced human-centered values
Behind sad news about CSCW comes great news about FAccT, as my solo-authored paper from my dissertation was accepted 🥳 I look forward to sharing more soon, but it's about the challenges that come with conducting ethnographic/human subjects work on the tech industry.
As you can see from my new name, I am on the job market this year! I am open to a variety of position types.
My research sits at the intersection of identity, technology design, and fairness/ethics.
I missed the wave of
#CHI2021
announcements, but I'm excited that this paper on inclusive web forms for non-binary people with
@aaroniidx
@katta_spiel
&
@whatknows
was accepted. We assessed inclusion, discomfort, and willingness to access services in different digital contexts.
Honored I was selected as a "young researcher" for the 11th Heidelberg Laureate Forum. The goal of the forum is to bring together junior scholars in math/CS with recipients of the Abel Prize, ACM A.M. Turing Award, ACM Prize in Computing, etc. Happy to rep HCI at this event 😎
I've uploaded a copy of our new
#CSCW2020
paper on race and gender categories in computer vision datasets! Authored with
@KandreaWade
@caitie_lustig
and
@whatknows
. Check out the full paper: Check out the blog post:
So a pretty cool thing happened - I was selected as a Presidential Appointment by the Colorado Senate for the Task Force for the Consideration of Facial Recognition Services. I am excited to learn what this kind of work entails & how I can make an impact in a political setting
Another preprint announcement! "A Framework of Severity for Harmful Content Online" by myself,
@aaroniidx
,
@cfiesler
,
@whatknows
to be published at CSCW 2021:
I have gotten a number of requests this week for fellowship tips! Finally got around to publishing a blog post on fellowship applications, alongside my materials (on my website and linked in the blog):
I am excited that a chapter of my dissertation work on the role of positional vantagepoints in the development of industrial scale computer vision got an accept to
#CHI2024
- look for "Products of Positionality: How Tech Workers Shape Identity Concepts in Computer Vision" soon :)
Person: What you working on?
Me: Pushing an academic paper through for a publication deadline this week.
Person: Oh dang, that's rough, hope you get a good grade.
💀
Honored to have received a Best Paper award for our
#CSCW2019
paper, How Computers See Gender. Be sure to come see my talk in Austin! It will be Nov 11 at the Gender, Identity, and Sexuality panel from 2:30 - 4:00
@math_rachel
Not to shamelessly self-plug but I'd love to add 3 papers on identity in AI:
1. How trans ppl perceive AGR:
2. How gender is classified in comm. face analysis:
3. How race & gender are categorized in CV lit:
I am unfortunately not at
#FAccT2024
this year, but I wanted to share my paper: "In the Walled Garden: Challenges and Opportunities for Research on the Practices of the AI Tech Industry."
I find the academic response to problematic Big Tech a bit troubling. Too often I see "just quit" or "stop selling your soul" on a Tweet an employee at Company X has posted about being troubled by their company. A thread on why this is not useful:
I have to admit, one of my favorite things about giving talks about gender binarism and trans representation in facial analysis/recognition technology is watching the talks about gender bias AFTER mine.
I just submitted my first NSF grant as a co-PI😳It was a massive learning experience for me and I'm so happy for the mentorship I've been getting here at CU towards my academic career. Now I'm planning to take what I learned and start a single PI grant!
I just realized how angry I am, like, on a personal level that people refuse to take COVID seriously. I've been diligently quarantining and social distancing for 105 days, yet since the collective U.S. has decided to just ignore COVID the cases continue to soar.
A note that I'm on the job market again this year! I'm primarily exploring tenure track options, as well as publication heavy industry and institute options. I love when people send me postings that they think would be relevant to me :)
Trying to be kind to myself during this time of uncertainty and stress - to my fellow job marketers, you're doing amazing and are fantastic worthy scholars 💖
"Accuracy is not the only judgment for ethical algorithmic decision making."
@benzevgreen
says of his award winning paper: much of ML decision making isn't evidenced based, despite being "data driven."
Finally released our SIGCHI Gender Inclusive Guidelines! Thank you to everyone who contributed! This is a living document so please feel free to offer feedback and share it with colleagues!
Now that rights review is finished, I've uploaded an official copy of our CSCW paper to my website! Check out "How Computers See Gender" by myself, Jacob Paul, and
@whatknows
for some more critical analysis of
#computervision
/
#FacialRecognition
/
#AGR
I made my very first tiktok video about the weird messages I get from straight dudes online.
Thank u to my queer crew for giving me the confidence to share this lmao.
Also don't send strangers weird DMs.
Happy sunday! I'm not sure this will be a useful thread for the majority of people following me, but I have seen some likes here and there that have prompted it. Plus, given all the discourse going on thx to JKR, I would like to discuss: transnormativity (SUPER LONG THREAD 1/21).
I want to highlight some frustrating things I regularly witness from well-meaning (usually cis) people regarding gender - particularly, feminine expression and its intersection with gender.
I got really excited that this face database paper I was analyzing mentioned ethics but apparently what they meant is it is the public's "ethical responsibility" to upload their images to the database 🙃
I just realized our new paper on facial analysis is officially in the ACM DL. Authored with
@KandreaWade
@caitie_lustig
&
@whatknows
Now y'all can officially cite it ;) (free copy on my website)
Presenting research that positions gender beyond a binary perspective of cis men and cis women, particularly to a pure CS crowd, has led to some particularly grating interactions for me already. Yet my work is increasingly recognized as not only valid, but necessary and inspiring
Today is the day,
#CSCW2019
! I'm presenting our award-winning paper, "How Computers See Gender," at 2:30 in Gender, Identity, and Sexuality in Salon C. Come check it out!
Had such a wonderful time at my first ever
@FAccTConference
! It was wonderful meeting many people I've only met online and many new people as well. I definitely want to connect further with many of you who I've had conversations with this week :)
Or (b) "We did this binary thing, which is binary, but I am so so sorry."
Or (c), which
@farbandish
apparently gets, people frantically trying to change their slides to reflect their binary selection as a limitation.
Frameworks for responsible / ethical / fair data collection are becoming more robust. What are some of your favorite (more recent) papers proposed for creating more ethical datasets for ML?
Privilege stems from how other people and systems see and classify you, and what the classification allows you to do or not do - not whether or not you identify with the privilege.
The attack on trans healthcare and abortion care and potentially marriage rights is also tied with current future of work debates. People are being trapped in states because of low wages. People are being trapped in states because companies refuse to continue remote work.
Happy Pride Month! Reminder that being fem in public sux. Regardless if you're a fem cis woman, trans woman, man, or non-binary person - going back into a physical office for 1 day reminded me how much anxiety fem people deal with regularly.
This round
#CSCW2021
reviews may be a time of stress and disappointment for some, or utter elation for others, but I want to express gratitude to reviewers. I really appreciate the critical reviews we got; they are going to be crucial in making our paper better :)
To close out our summer speaker series, we heard last week from
@morganklauss
about his research on "How Computers See Gender" - what are the implications of binary gender classification, particularly for people who don't fit neatly into the categories defined by the system?
RE: Fellowships - I am also wondering if it would be useful for people if I put my application materials on my website. I doubt there is a particular formula for these things, but maybe it could be helpful? Does
#AcademicTwitter
have thoughts on this?
Every time I do a lit review of any sort, I tell myself, "I should really re-organize my Zotero" but the thought of how long this will take is horrifying
My
#CSCW
paper was... rejected! Whoops. We had really amazing reviews nonetheless. I'm excited to work on it more and resubmit (or submit elsewhere) :)
@timnitGebru
This has been a salient feeling w/ students in my department, particularly women of color. We actually decided to have a student meeting about feeling demoralized about the future of ethics research. It really sent a message from industry to young researchers.
Today is
#ShutDownSTEM
and
#Strike4BlackLives
. I encourage my fellow white people in STEM to use this time to engage with the racism in their fields. Share what you're planning to do below:
Being at
#Neurips2019
has been fascinating for me as an HCI scholar. Beyond meeting awesome people outside my discipline, I'm learning far more about the epistemological thinking/approaches of ML-driven comp sci as a field than I had expected.
This is huge. The ACM is calling for an "immediate suspension of the current and future private and governmental use" of Facial Recognition technologies, for "both technical and ethical reasons."
As I've watched this happen over the last few days I've debated even saying something vague. But there are even students in the whisper network with documented accusations against them that have not been disciplined by faculty and industry partners who know.
In this paper, we extend prior arguements that facial analysis technology is inherently physiognomic. We dig more deeply into the gendered histories of physiognomy & colonialism and map these histories to how gender has permeated modern technological tools like in computer vision
One extremely salient microaggression towards trans people is the focus on "attractiveness." From "but you are so beautiful as a girl, what a waste" to "you're hot as a woman so you'll be an attractive man," there is an obsession with in/validating transness thru sexualization.
When you write an entire paper, feel like it could be better, so you basically take the whole thing apart and now it's a jumbled mess of paragraphs and now you can do nothing but stare blankly at it for an hour.