Efficient and Practical Approximation Algorithms for Advertising in Content Feeds

2025-02-04

AI Summary

This paper addresses the constrained native advertising insertion problem in content-streaming platforms (e.g., X, TikTok), where ads may only be inserted immediately after semantically relevant organic posts, and user attention decays dynamically with sequential content and ad consumption. The objective is to maximize engagement-based rewards (e.g., clicks) under sequence-level placement constraints and time-varying attention decay. To this end, we propose the first practical 2-approximation greedy algorithm. Its key innovation lies in identifying and leveraging a “tail-content attention” effect—where users disproportionately attend to content near the stream’s end—to strictly improve the approximation guarantee. We also introduce the first systematic empirical evaluation framework for this problem. Our algorithm is computationally efficient, deployment-ready, and outperforms existing 4-approximation methods significantly on real-world data, achieving superior ad performance while preserving user experience.

Technology Category

• Search and Optimization: Algorithm Configuration
• Constraint Satisfaction and Optimization: Distributed CSP/Optimization
• Machine Learning: Online Learning & Bandits

Application Category

• User Modeling, Personalization and Recommendation: User modeling for targeted and personalized online advertising
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web
• Web Mining and Content Analysis: Content-based information diffusion

Abstract

Content feeds provided by platforms such as X (formerly Twitter) and TikTok are consumed by users on a daily basis. In this paper, we revisit the native advertising problem in content feeds, initiated by Ieong et al. Given a sequence of organic items (e.g., videos or posts) relevant to a user's interests or to an information search, the goal is to place ads within the organic content so as to maximize a reward function (e.g., number of clicks), while accounting for two considerations: (1) an ad can only be inserted after a relevant content item; (2) the users' attention decays after consuming content or ads. These considerations provide a natural model for capturing both the advertisement effectiveness and the user experience. In this paper, we design fast and practical 2-approximation greedy algorithms for the associated optimization problem, improving over the best-known practical algorithm that only achieves an approximation factor of~4. Our algorithms exploit a counter-intuitive observation, namely, while top items are seemingly more important due to the decaying attention of the user, taking good care of the bottom items is key for obtaining improved approximation guarantees. We then provide the first comprehensive empirical evaluation on the problem, showing the strong empirical performance of our~methods.

Problem

• Optimize ad placement in content feeds
• Maximize reward function for ad clicks
• Address user attention decay in content consumption

Innovation

• 2-approximation greedy algorithms
• decaying user attention model
• comprehensive empirical evaluation