algorithmic choice & control
In addition, curation tools should enable easy iteration to improve initial creations incrementally, as so-cial media users may be more amenable to short tasks spread out over many sessions as opposed to s single lengthy task. Instead, existing systems often assume that users want to create highly performant custom classifiers in a single sitting [... See more
End User Authoring of Personalized Content Classifiers: Comparing Example Labeling, Rule Writing, and LLM Prompting
Threads Tests New Way to Manually Guide Feed Algorithms
socialmediatoday.com
Bluesky "For You" feed playground
linklonk.comConnective Posts: A New Approach to Political Conversations on Social Media
mediaengagement.org
The findings suggest that people can improve their online discussions by using this type of language in their posts. It also suggests social media platforms should optimize algorithms to surface these posts more often...Center for Media Engagement research offers a solution to this problem—encouraging the use of “connective language” online.3 Connective language expresses an openness to other perspectives and a humble approach to presenting one’s viewpoints.4 Examples of connective language include phrases such as “correct me if I’m wrong” or “this is just my opinion.”
Top feeds heavily skew towards English content, although there were five Japanese,
four Portuguese, and a single feed each for Indonesian, Ukrainian, and Spanish. The
relative prominence of Japanese and Portuguese reflects the geographic distribution of
the platform’s users (Duarte, 2025). In terms of topic, identity-focused feeds were most
prominent,... See more
four Portuguese, and a single feed each for Indonesian, Ukrainian, and Spanish. The
relative prominence of Japanese and Portuguese reflects the geographic distribution of
the platform’s users (Duarte, 2025). In terms of topic, identity-focused feeds were most
prominent,... See more
However, many tools for authoring personal classifiers on so-cial media are designed for users with heightened motivation , such as community moderators [ 12 , 40 ] and high-profile content cre-ators [ 41 ]. In focusing on these users, such tools neglect the usage patterns of general social media users, who spend significant time on social media... See more
End User Authoring of Personalized Content Classifiers: Comparing Example Labeling, Rule Writing, and LLM Prompting
Bringing Better Feeds to Life – Knight-Georgetown Institute
kgi.georgetown.edu
To support end-user customization, researchers have examined and built a variety of specialized tools that enable users to author person-alized content classifiers within social media [35, 40, 41, 63]. They have also explored generic techniques for non-technical people to build their own text classifiers