Multimodal Graph Retrieval-Augmented Sequential Recommendation via Collaborative Filtering Paths

Key points

• Multimodal Large Language Models (MLLMs) have demonstrated strong potential for sequential recommendation through their ability to reason over complex multimodal data.1
• However, existing approaches either rely solely on the target user's own interaction history, neglecting collaborative signals from neighboring users, or incur substantial computational overhead through repeated MLLM inference over long interaction histories.1
• MGRASRec injects collaborative filtering signals conditioned on the candidate item directly into the MLLM prompt by retrieving structured paths from a user-item interaction graph, extended via multimodal similarity to increase coverage beyond exact co-interaction overlap.1
• This retrieval also surfaces the history items most relevant to the candidate at no additional cost, removing the need for recurrent summarization and keeping inference to a single forward pass per candidate.1

Sources (1)

• [1]Multimodal Graph Retrieval-Augmented Sequential Recommendation via Collaborative Filtering Paths
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 8, 04:27 AM

  To address these challenges, we propose MGRASRec, a multimodal graph retrieval-augmented framework for sequential recommendation.

  Multimodal Large Language Models (MLLMs) have demonstrated strong potential for sequential recommendation through their ability to reason over complex multimodal data.

Extractive summary: sentences quoted from the sources.