Facebook AI tools are giving users more control over the content they see in Feed and Reels by allowing them to describe their interests in their own words. The new features let users customize recommendations and adjust what appears in their feeds based on their preferences.

The changes introduce a feature called Your algorithm, along with new feedback options that allow users to fine-tune recommendations. According to Facebook, these tools use artificial intelligence (AI) to make content recommendations more relevant to individual interests.

The platform has also updated the three-dot menu on posts, making it easier for users to access content customization options.

Your Algorithm Lets Users Set Their Interests

Facebook is rolling out Your algorithm, a dedicated space where users can add and manage the topics they want to see in Feed and Reels.

Accessible through the three-dot menu, the feature allows users to communicate their interests directly rather than relying entirely on the platform to infer their preferences from their activity.

As a result, users can provide more specific information about the content they want to see, helping Facebook tailor recommendations to their interests.

New Feedback Options Refine Recommendations

Facebook is also expanding the ways users can provide feedback on individual posts in Feed and Reels.

When users select ‘Interested’ or ‘Not interested’, the platform may ask follow-up questions to understand their preferences in greater detail. Users can then describe, in their own words, the content they would like to see more or less often.

According to Facebook, recommendations can update in real time based on this feedback. This allows users to adjust their experience more directly as their interests and preferences change.

The updated three-dot menu also makes these customization tools easier to access while browsing content.

Large Language Models Support Content Ranking

Facebook’s existing ranking systems use multiple signals to determine which content users see. These include likes, comments, viewing activity and other forms of engagement.

The platform is now incorporating large language models (LLMs) into this process to better understand users’ interests, context and intent when they express preferences in natural language.

Unlike feedback based solely on engagement patterns, natural language input allows users to explain more precisely what they want to see. The system can then combine this information with existing signals to refine recommendations.

Facebook says its ranking systems can adapt within the same session, allowing Feed and Reels recommendations to reflect users’ stated preferences more quickly.

More Direct Control Over Facebook Recommendations

The new features represent a shift towards more interactive content personalization, giving users additional ways to influence how Facebook’s recommendation systems operate.

By combining natural language feedback with existing engagement signals, Facebook AI tools aim to make Feed and Reels recommendations more responsive to individual interests.

With Your algorithm and the expanded feedback options, Facebook is giving users a more direct way to shape the content they see, rather than relying exclusively on recommendations inferred from their activity.

Related post

View all