Recommender systems can learn from how r\ifyoulikeblank users solicit recommendations from each other. Our @ACM_CSCW paper led by @momen_bh shows unique recommendation forms (e.g., a movie that captures Lofi hip-hop vibe) w\ @Donghan9508, Jelson, @tanmit https://t.co/nZvS3BYbQN
@sama could you train your AI not to vomit 5-paragraphs every time my students ask it to write an email about XYZ? If I have to read those, how will I procrastinate 10hours a day?
Distressing news!! @YouTube reverses stance on misinfo, allowing elec denialism. In our @acm_chi paper, we found positive impact of their prev policy with fewer elect fraud videos in search results. This decision is a significant step backward. Our paper https://t.co/jJKeVn6KbP
Our study reveals that users have better precision in identifying similarity than identifying disparity. It also shows some impact on their credibility perception when an article pair has high contrast. See the full paper here: https://t.co/2V2UAQRDnM
Are you frustrated by media bias and wonder if you could simultaneously engage with news articles from opposing-leaning sources? Our work at #CHI2023 w/ @tanmit@snaglee2401@niteshg does just that with a comparative news annotation interface NewsComp https://t.co/kUzqYNT71f
We tested users' performance in annotating similarities & important disparities between news article pairs on contentious topics like abortion and whether such engagement impacts their attitude toward the articles
@im__jane Yeah absolutely. I was looking forward to speaking to you and other familiar faces. I even had plans with @Prerna_99 after the conference. Unfortunately USCIS happened. Hope you folks get to enjoy.
New on PNAS NEXUS 🚨
*Deplatforming Parler following January 6th did not decrease the activity of Parler users in fringe social media in general* Users migrated to other platforms like Gab at a rate that compensated for the decrease in activity in Parler
📰https://t.co/XlaGHLi0BI
RT! Echolab is seeking people with experience using IFTTT pro+ for an interview and user study. If you’re interested in participating, please fill out the form here: https://t.co/gMX1uUnVsk The participants will be compensated (e-gift $60) card.
@ndiakopoulos Also, the column was written by Bret Stephens, the guy whose writings are likely to be based on false evidence (https://t.co/fGSIhHj9g7 dissecting his "The Secrets of Jewish Genius")
@jonathanstray Agree: need better agent modeling. Disagree: junk (eg, in both simulation extreme recs still increases and people watches whats in front of them). Great work @manoelribeiro. wondering, does the simulation capture YT tend to show same recs over&over and if that impacts outcome?
On a personal note, I worked on my #CHI2023 paper during a tough time in my life and I am proud of myself for completing the work and successfully submitting it to the conference. And I got lucky that the paper got accepted 😊
My paper - "Assessing enactment of content regulation policies: A post hoc crowd-sourced audit of election misinformation on YouTube" with @momen_bh and @tanmit got accepted to #CHI2023.
@uw_ischool@uwcip@VT_CS (1/6)
I will be presenting this work at the #CSCW2022 Misinformation, Fake News & Conspiracy Theories sessions tomorrow at 8:40pm PST and next Wednesday at 8:40am PST. Come say hi!