# Compass: Continuously Aligning Social Media Feeds via In-Situ Reflections

arXiv: https://arxiv.org/abs/2608.04274 · HTML: https://mlsanity.com/paper/2608.04274/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 26 of its 90 references that are in the map.
- First posted on arXiv 4 Aug 2026.

## Builds on (most cited first, 26 of 26)

- [The Vendi Score: A Diversity Evaluation Metric for Machine Learning](https://mlsanity.com/paper/2210.02410/) (2022)
- [RAPPOR: Randomized Aggregatable Privacy-Preserving Ordinal Response](https://mlsanity.com/paper/1407.6981/) (2014)
- [Everyday algorithm auditing: Understanding the power of everyday users in surfacing harmful algorithmic behaviors](https://mlsanity.com/paper/2105.02980/) (2021)
- [The Challenge of Understanding What Users Want: Inconsistent Preferences and Engagement Optimization](https://mlsanity.com/paper/2202.11776/) (2022)
- [Estimating and Penalizing Induced Preference Shifts in Recommender Systems](https://mlsanity.com/paper/2204.11966/) (2022)
- [Embedding Democratic Values into Social Media AIs via Societal Objective Functions](https://mlsanity.com/paper/2307.13912/) (2023)
- [Personalizing Content Moderation on Social Media: User Perspectives on Moderation Choices, Interface Design, and Labor](https://mlsanity.com/paper/2305.10374/) (2023)
- [Watch Your Language: Investigating Content Moderation with Large Language Models](https://mlsanity.com/paper/2309.14517/) (2023)
- [How the Design of YouTube Influences User Sense of Agency](https://mlsanity.com/paper/2101.11778/) (2021)
- [Interactive Recommendation Agent with Active User Commands](https://mlsanity.com/paper/2509.21317/) (2025)
- [Analyzing User Engagement with TikTok's Short Format Video Recommendations using Data Donations](https://mlsanity.com/paper/2301.04945/) (2023)
- [End User Authoring of Personalized Content Classifiers: Comparing Example Labeling, Rule Writing, and LLM Prompting](https://mlsanity.com/paper/2409.03247/) (2024)
- [TikTok and the Art of Personalization: Investigating Exploration and Exploitation on Social Media Feeds](https://mlsanity.com/paper/2403.12410/) (2024)
- [What We Know About Using Non-Engagement Signals in Content Ranking](https://mlsanity.com/paper/2402.06831/) (2024)
- [Bluesky: Network Topology, Polarization, and Algorithmic Curation](https://mlsanity.com/paper/2405.17571/) (2024)
- [Bonsai: Intentional and Personalized Social Media Feeds](https://mlsanity.com/paper/2509.10776/) (2025)
- [Mapping the Design Space of Teachable Social Media Feed Experiences](https://mlsanity.com/paper/2401.14000/) (2024)
- [Alexandria: A Library of Pluralistic Values for Realtime Re-Ranking of Social Media Feeds](https://mlsanity.com/paper/2505.10839/) (2025)
- [Reranking Social Media Feeds: A Practical Guide for Field Experiments](https://mlsanity.com/paper/2406.19571/) (2024)
- [Studying Behavioral Addiction by Combining Surveys and Digital Traces: A Case Study of TikTok](https://mlsanity.com/paper/2501.15539/) (2025)
- [CTRL-Rec: Controlling Recommender Systems With Natural Language](https://mlsanity.com/paper/2510.12742/) (2025)
- [Beyond Explicit and Implicit: How Users Provide Feedback to Shape Personalized Recommendation Content](https://mlsanity.com/paper/2502.09869/) (2025)
- [Ethical User Interfaces: Exploring the Effects of Dark Patterns on Facebook](https://mlsanity.com/paper/2104.03010/) (2021)
- [Understanding Decentralized Social Feed Curation on Mastodon](https://mlsanity.com/paper/2504.18817/) (2025)
- [PILOT: Control Surfaces for Authoring Social Media Feeds](https://mlsanity.com/paper/2509.19615/) (2025)
- ["They've Over-Emphasized That One Search": Controlling Unwanted Content on TikTok's For You Page](https://mlsanity.com/paper/2504.13895/) (2025)

## Abstract (arXiv)

Social media recommendation feeds often optimize for users' immediate impulses rather than preferences they would hold after deeper reflection. Some systems address this misalignment by incorporating users' explicit preferences via a configuration page or in-feed controls instead of just behavioral signals. However, users typically have evolving preferences, and their stated preferences and behavior naturally diverge, necessitating continuous reflection and feed realignment. But existing strategies require the user to take initiative and are often effortful; as a result, in practice they are rarely invoked. We present Compass, a system that aligns a user's feed with their reflective preferences by helping users reflect on and articulate their preferences given their behavior. To enable continuous reflection during everyday browsing, Compass surfaces in-situ reflections via lightweight notifications, while feed alignment is achieved by periodically simulating behavioral signals and directly manipulating feed content. We embedded Compass within YouTube Shorts and compared it against a baseline without continuous support through a 10-day field study (N=15). We found that Compass promoted more reflective and purposeful feed consumption, iterative preference adjustment, and stronger feed alignment, without sacrificing the casual nature of feed browsing.