Let It Go? Not Quite: Addressing Item Cold Start in Sequential Recommendations with Content-Based Initialization

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Pembek, Anton, Fatkulin, Artem, Klenitskiy, Anton, Vasilev, Alexey
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866909705438232576
author Pembek, Anton
Fatkulin, Artem
Klenitskiy, Anton
Vasilev, Alexey
author_facet Pembek, Anton
Fatkulin, Artem
Klenitskiy, Anton
Vasilev, Alexey
contents Many sequential recommender systems suffer from the cold start problem, where items with few or no interactions cannot be effectively used by the model due to the absence of a trained embedding. Content-based approaches, which leverage item metadata, are commonly used in such scenarios. One possible way is to use embeddings derived from content features such as textual descriptions as initialization for the model embeddings. However, directly using frozen content embeddings often results in suboptimal performance, as they may not fully adapt to the recommendation task. On the other hand, fine-tuning these embeddings can degrade performance for cold-start items, as item representations may drift far from their original structure after training. We propose a novel approach to address this limitation. Instead of entirely freezing the content embeddings or fine-tuning them extensively, we introduce a small trainable delta to frozen embeddings that enables the model to adapt item representations without letting them go too far from their original semantic structure. This approach demonstrates consistent improvements across multiple datasets and modalities, including e-commerce datasets with textual descriptions and a music dataset with audio-based representation.
format Preprint
id arxiv_https___arxiv_org_abs_2507_19473
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Let It Go? Not Quite: Addressing Item Cold Start in Sequential Recommendations with Content-Based Initialization
Pembek, Anton
Fatkulin, Artem
Klenitskiy, Anton
Vasilev, Alexey
Information Retrieval
Artificial Intelligence
Machine Learning
Many sequential recommender systems suffer from the cold start problem, where items with few or no interactions cannot be effectively used by the model due to the absence of a trained embedding. Content-based approaches, which leverage item metadata, are commonly used in such scenarios. One possible way is to use embeddings derived from content features such as textual descriptions as initialization for the model embeddings. However, directly using frozen content embeddings often results in suboptimal performance, as they may not fully adapt to the recommendation task. On the other hand, fine-tuning these embeddings can degrade performance for cold-start items, as item representations may drift far from their original structure after training. We propose a novel approach to address this limitation. Instead of entirely freezing the content embeddings or fine-tuning them extensively, we introduce a small trainable delta to frozen embeddings that enables the model to adapt item representations without letting them go too far from their original semantic structure. This approach demonstrates consistent improvements across multiple datasets and modalities, including e-commerce datasets with textual descriptions and a music dataset with audio-based representation.
title Let It Go? Not Quite: Addressing Item Cold Start in Sequential Recommendations with Content-Based Initialization
topic Information Retrieval
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2507.19473