FACE: A General Framework for Mapping Collaborative Filtering Embeddings into LLM Tokens

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Wang, Chao, Song, Yixin, Ye, Jinhui, Qin, Chuan, Shen, Dazhong, Liu, Lingfeng, Wang, Xiang, Zhang, Yanyong
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912655010168832
author Wang, Chao
Song, Yixin
Ye, Jinhui
Qin, Chuan
Shen, Dazhong
Liu, Lingfeng
Wang, Xiang
Zhang, Yanyong
author_facet Wang, Chao
Song, Yixin
Ye, Jinhui
Qin, Chuan
Shen, Dazhong
Liu, Lingfeng
Wang, Xiang
Zhang, Yanyong
contents Recently, large language models (LLMs) have been explored for integration with collaborative filtering (CF)-based recommendation systems, which are crucial for personalizing user experiences. However, a key challenge is that LLMs struggle to interpret the latent, non-semantic embeddings produced by CF approaches, limiting recommendation effectiveness and further applications. To address this, we propose FACE, a general interpretable framework that maps CF embeddings into pre-trained LLM tokens. Specifically, we introduce a disentangled projection module to decompose CF embeddings into concept-specific vectors, followed by a quantized autoencoder to convert continuous embeddings into LLM tokens (descriptors). Then, we design a contrastive alignment objective to ensure that the tokens align with corresponding textual signals. Hence, the model-agnostic FACE framework achieves semantic alignment without fine-tuning LLMs and enhances recommendation performance by leveraging their pre-trained capabilities. Empirical results on three real-world recommendation datasets demonstrate performance improvements in benchmark models, with interpretability studies confirming the interpretability of the descriptors. Code is available in https://github.com/YixinRoll/FACE.
format Preprint
id arxiv_https___arxiv_org_abs_2510_15729
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FACE: A General Framework for Mapping Collaborative Filtering Embeddings into LLM Tokens
Wang, Chao
Song, Yixin
Ye, Jinhui
Qin, Chuan
Shen, Dazhong
Liu, Lingfeng
Wang, Xiang
Zhang, Yanyong
Information Retrieval
Recently, large language models (LLMs) have been explored for integration with collaborative filtering (CF)-based recommendation systems, which are crucial for personalizing user experiences. However, a key challenge is that LLMs struggle to interpret the latent, non-semantic embeddings produced by CF approaches, limiting recommendation effectiveness and further applications. To address this, we propose FACE, a general interpretable framework that maps CF embeddings into pre-trained LLM tokens. Specifically, we introduce a disentangled projection module to decompose CF embeddings into concept-specific vectors, followed by a quantized autoencoder to convert continuous embeddings into LLM tokens (descriptors). Then, we design a contrastive alignment objective to ensure that the tokens align with corresponding textual signals. Hence, the model-agnostic FACE framework achieves semantic alignment without fine-tuning LLMs and enhances recommendation performance by leveraging their pre-trained capabilities. Empirical results on three real-world recommendation datasets demonstrate performance improvements in benchmark models, with interpretability studies confirming the interpretability of the descriptors. Code is available in https://github.com/YixinRoll/FACE.
title FACE: A General Framework for Mapping Collaborative Filtering Embeddings into LLM Tokens
topic Information Retrieval
url https://arxiv.org/abs/2510.15729