GRAM: Generative Recommendation via Semantic-aware Multi-granular Late Fusion
This work improves recommendation systems for users by enhancing generative models with semantic-aware techniques, though it is incremental as it builds on existing generative recommendation paradigms.
The paper tackles the problem of generative recommendation by addressing limitations in incorporating implicit item relationships and utilizing lengthy item information, resulting in a model that outperforms eight state-of-the-art models with improvements of 11.5-16.0% in Recall@5 and 5.3-13.6% in NDCG@5.
Generative recommendation is an emerging paradigm that leverages the extensive knowledge of large language models by formulating recommendations into a text-to-text generation task. However, existing studies face two key limitations in (i) incorporating implicit item relationships and (ii) utilizing rich yet lengthy item information. To address these challenges, we propose a Generative Recommender via semantic-Aware Multi-granular late fusion (GRAM), introducing two synergistic innovations. First, we design semantic-to-lexical translation to encode implicit hierarchical and collaborative item relationships into the vocabulary space of LLMs. Second, we present multi-granular late fusion to integrate rich semantics efficiently with minimal information loss. It employs separate encoders for multi-granular prompts, delaying the fusion until the decoding stage. Experiments on four benchmark datasets show that GRAM outperforms eight state-of-the-art generative recommendation models, achieving significant improvements of 11.5-16.0% in Recall@5 and 5.3-13.6% in NDCG@5. The source code is available at https://github.com/skleee/GRAM.