EAGER: Two-Stream Generative Recommender with Behavior-Semantic Collaboration

TL;DR

EAGER is a new generative recommendation framework that effectively combines behavioral and semantic information using a two-stream architecture, leading to improved performance on benchmark datasets.

Contribution

It introduces a novel two-stream generative architecture with specialized decoders and contrastive tasks to integrate behavioral and semantic data in recommendations.

Findings

Outperforms existing methods on four public benchmarks.

Effectively models both behavioral and semantic item features.

Demonstrates the benefit of cross-interaction between feature types.

Abstract

Generative retrieval has recently emerged as a promising approach to sequential recommendation, framing candidate item retrieval as an autoregressive sequence generation problem. However, existing generative methods typically focus solely on either behavioral or semantic aspects of item information, neglecting their complementary nature and thus resulting in limited effectiveness. To address this limitation, we introduce EAGER, a novel generative recommendation framework that seamlessly integrates both behavioral and semantic information. Specifically, we identify three key challenges in combining these two types of information: a unified generative architecture capable of handling two feature types, ensuring sufficient and independent learning for each type, and fostering subtle interactions that enhance collaborative information utilization. To achieve these goals, we propose (1) a two-stream generation architecture leveraging a shared encoder and two separate decoders to decode behavior tokens and semantic tokens with a confidence-based ranking strategy; (2) a global contrastive task with summary tokens to achieve discriminative decoding for each type of information; and (3) a semantic-guided transfer task designed to implicitly promote cross-interactions through reconstruction and estimation objectives. We validate the effectiveness of EAGER on four public benchmarks, demonstrating its superior performance compared to existing methods.