Alexandria: A Library of Pluralistic Values for Realtime Re-Ranking of Social Media Feeds

TL;DR

This paper introduces Alexandria, a library of 78 pluralistic values for social media feed re-ranking, implemented via LLM classifiers, enabling real-time, user-controlled feed customization to better align with diverse user needs.

Contribution

The paper presents a comprehensive library of social media values and a practical tool for real-time, user-driven feed re-ranking using LLM classifiers, addressing limitations of engagement-focused algorithms.

Findings

Diverse user needs benefit from a large, nuanced value library.

User control improves satisfaction and alignment with individual preferences.

The tool effectively operationalizes pluralistic values in real-time feed re-ranking.

Abstract

Social media feed ranking algorithms fail when they too narrowly focus on engagement as their objective. The literature has asserted a wide variety of values that these algorithms should account for as well -- ranging from well-being to productive discourse -- far more than can be encapsulated by a single topic or theory. In response, we present a $\textit{library of values}$ for social media algorithms: a pluralistic set of 78 values as articulated across the literature, implemented into LLM-powered content classifiers that can be installed individually or in combination for real-time re-ranking of social media feeds. We investigate this approach by developing a browser extension, $\textit{Alexandria}$, that re-ranks the X/Twitter feed in real time based on the user's desired values. Through two user studies, both qualitative (N=12) and quantitative (N=257), we found that diverse user needs require a large library of values, enabling more nuanced preferences and greater user control. With this work, we argue that the values criticized as missing from social media ranking algorithms can be operationalized and deployed today through end-user tools.