Controlling Diversity at Inference: Guiding Diffusion Recommender Models with Targeted Category Preferences

TL;DR

This paper introduces D3Rec, a diffusion-based recommender system that allows flexible control over diversity and category preferences during inference, addressing limitations of previous static methods.

Contribution

D3Rec is a novel end-to-end diffusion model that enables dynamic diversity control and targeted category preferences during inference in recommender systems.

Findings

D3Rec effectively controls diversity at inference in real-world datasets.

The model adapts to arbitrary targeted category preferences.

Experimental results show improved diversity and accuracy trade-offs.

Abstract

Diversity control is an important task to alleviate bias amplification and filter bubble problems. The desired degree of diversity may fluctuate based on users' daily moods or business strategies. However, existing methods for controlling diversity often lack flexibility, as diversity is decided during training and cannot be easily modified during inference. We propose \textbf{D3Rec} (\underline{D}isentangled \underline{D}iffusion model for \underline{D}iversified \underline{Rec}ommendation), an end-to-end method that controls the accuracy-diversity trade-off at inference. D3Rec meets our three desiderata by (1) generating recommendations based on category preferences, (2) controlling category preferences during the inference phase, and (3) adapting to arbitrary targeted category preferences. In the forward process, D3Rec removes category preferences lurking in user interactions by adding noises. Then, in the reverse process, D3Rec generates recommendations through denoising steps while reflecting desired category preferences. Extensive experiments on real-world and synthetic datasets validate the effectiveness of D3Rec in controlling diversity at inference.