Enhancing High-order Interaction Awareness in LLM-based Recommender Model
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866929594656882688 |
|---|---|
| author | Wang, Xinfeng Cui, Jin Fukumoto, Fumiyo Suzuki, Yoshimi |
| author_facet | Wang, Xinfeng Cui, Jin Fukumoto, Fumiyo Suzuki, Yoshimi |
| contents | Large language models (LLMs) have demonstrated prominent reasoning capabilities in recommendation tasks by transforming them into text-generation tasks. However, existing approaches either disregard or ineffectively model the user-item high-order interactions. To this end, this paper presents an enhanced LLM-based recommender (ELMRec). We enhance whole-word embeddings to substantially enhance LLMs' interpretation of graph-constructed interactions for recommendations, without requiring graph pre-training. This finding may inspire endeavors to incorporate rich knowledge graphs into LLM-based recommenders via whole-word embedding. We also found that LLMs often recommend items based on users' earlier interactions rather than recent ones, and present a reranking solution. Our ELMRec outperforms state-of-the-art (SOTA) methods in both direct and sequential recommendations. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_19979 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Enhancing High-order Interaction Awareness in LLM-based Recommender Model Wang, Xinfeng Cui, Jin Fukumoto, Fumiyo Suzuki, Yoshimi Information Retrieval Computation and Language Large language models (LLMs) have demonstrated prominent reasoning capabilities in recommendation tasks by transforming them into text-generation tasks. However, existing approaches either disregard or ineffectively model the user-item high-order interactions. To this end, this paper presents an enhanced LLM-based recommender (ELMRec). We enhance whole-word embeddings to substantially enhance LLMs' interpretation of graph-constructed interactions for recommendations, without requiring graph pre-training. This finding may inspire endeavors to incorporate rich knowledge graphs into LLM-based recommenders via whole-word embedding. We also found that LLMs often recommend items based on users' earlier interactions rather than recent ones, and present a reranking solution. Our ELMRec outperforms state-of-the-art (SOTA) methods in both direct and sequential recommendations. |
| title | Enhancing High-order Interaction Awareness in LLM-based Recommender Model |
| topic | Information Retrieval Computation and Language |
| url | https://arxiv.org/abs/2409.19979 |