MARC: Multimodal and Multi-Task Agentic Retrieval-Augmented Generation for Cold-Start Recommender System

TL;DR

MARC is a novel multimodal, multi-task recommender system leveraging agentic retrieval-augmented generation and graph databases to improve cold-start recommendations in the cocktail domain, demonstrating superior answer quality.

Contribution

The paper introduces MARC, combining multimodal data, multi-task learning, and agentic RAG with graph databases for cold-start recommender systems, specifically in the cocktail domain.

Findings

Graph database-based answers outperformed vector database in quality.

System effectively generates contextually appropriate responses.

Evaluation with both LLM judges and humans confirms improved performance.

Abstract

Recommender systems (RS) are currently being studied to mitigate limitations during cold-start conditions by leveraging modality information or introducing Agent concepts based on the exceptional reasoning capabilities of Large Language Models (LLMs). Meanwhile, food and beverage recommender systems have traditionally used knowledge graph and ontology concepts due to the domain's unique data attributes and relationship characteristics. On this background, we propose MARC, a multimodal and multi-task cocktail recommender system based on Agentic Retrieval-Augmented Generation (RAG) utilizing graph database under cold-start conditions. The proposed system generates high-quality, contextually appropriate answers through two core processes: a task recognition router and a reflection process. The graph database was constructed by processing cocktail data from Kaggle, and its effectiveness was evaluated using 200 manually crafted questions. The evaluation used both LLM-as-a-judge and human evaluation to demonstrate that answers generated via the graph database outperformed those from a simple vector database in terms of quality. The code is available at https://github.com/diddbwls/cocktail_rec_agentrag