LLM-based Fusion of Multi-modal Features for Commercial Memorability Prediction

Fuente: arXiv
Saved in:
Bibliographic Details
Main Author: Pramov, Aleksandar
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866909870794473472
author Pramov, Aleksandar
author_facet Pramov, Aleksandar
contents This paper addresses the prediction of commercial (brand) memorability as part of "Subtask 2: Commercial/Ad Memorability" within the "Memorability: Predicting movie and commercial memorability" task at the MediaEval 2025 workshop competition. We propose a multimodal fusion system with a Gemma-3 LLM backbone that integrates pre-computed visual (ViT) and textual (E5) features by multi-modal projections. The model is adapted using Low-Rank Adaptation (LoRA). A heavily-tuned ensemble of gradient boosted trees serves as a baseline. A key contribution is the use of LLM-generated rationale prompts, grounded in expert-derived aspects of memorability, to guide the fusion model. The results demonstrate that the LLM-based system exhibits greater robustness and generalization performance on the final test set, compared to the baseline. The paper's codebase can be found at https://github.com/dsgt-arc/mediaeval-2025-memorability
format Preprint
id arxiv_https___arxiv_org_abs_2510_22829
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LLM-based Fusion of Multi-modal Features for Commercial Memorability Prediction
Pramov, Aleksandar
Computer Vision and Pattern Recognition
Artificial Intelligence
Multimedia
This paper addresses the prediction of commercial (brand) memorability as part of "Subtask 2: Commercial/Ad Memorability" within the "Memorability: Predicting movie and commercial memorability" task at the MediaEval 2025 workshop competition. We propose a multimodal fusion system with a Gemma-3 LLM backbone that integrates pre-computed visual (ViT) and textual (E5) features by multi-modal projections. The model is adapted using Low-Rank Adaptation (LoRA). A heavily-tuned ensemble of gradient boosted trees serves as a baseline. A key contribution is the use of LLM-generated rationale prompts, grounded in expert-derived aspects of memorability, to guide the fusion model. The results demonstrate that the LLM-based system exhibits greater robustness and generalization performance on the final test set, compared to the baseline. The paper's codebase can be found at https://github.com/dsgt-arc/mediaeval-2025-memorability
title LLM-based Fusion of Multi-modal Features for Commercial Memorability Prediction
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Multimedia
url https://arxiv.org/abs/2510.22829