Jaime Bellver-Soler

Jaime Bellver-Soler

Auto-generated profile, based on publicly available information

Jaime Bellver-Soler

THAU Group, IPTC, Universidad Politécnica de Madrid, Madrid, Spain

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 October 2024
Verified
Conference Paper
Open access
IMIván Martín-Fernández
Iván Martín-Fernández
SESergio Esteban-Romero
Sergio Esteban-Romero
JBJaime Bellver-Soler
Jaime Bellver-Soler
Fernando Fernández-MartínezFernando Fernández-Martínez
Fernando Fernández-Martínez
Manuel Gil-MartínManuel Gil-Martín
Manuel Gil-Martín
Designing reliable automatic models for social perception can contribute to a better understanding of human behavior, enabling more trustworthy experiences in the multimedia on-line communication environment. However, predicting social attributes from video data remains challenging due to the complex interplay of visual, auditory, and linguistic cues. In this paper, we address this challenge by investigating the effectiveness of Multimodal Large Language Models (MM-LLMs) for feature extraction in the MuSe-Perception challenge. Firstly, our analysis of the novel LMU-ELP dataset has revealed high correlations between certain perceptual dimensions, motivating using a single regression model for all 16 social attributes to be predicted for a set of speakers appearing in recorded video clips. We demonstrate that dimensionality reduction through Principal Component Analysis (PCA) can be applied to the label space without a relevant performance loss. Secondly, by employing frozen MM-LLMs as feature extractors, we explore their ability to capture perception-related information. We extract sequence embeddings from the Qwen-VL and Qwen-Audio models and train a Multi-Layer Perceptron over the attention-pooled vectors for each one of the encoders, obtaining a mean Pearson correlation of 0.22 using the average predictions for both models. Our best result of 0.31 is achieved by training the same architecture over the baseline vit-ver and w2v-msp features, which motivates further exploration on how to effectively leverage advanced MM-LLMs as feature extractors. Lastly, a post hoc analysis of our results highlights the limitations of Pearson correlation for evaluating regression performance in this context. In particular, a similar Pearson coefficient can be obtained with two very different prediction sets displaying different levels of variability. We take this result as a call to action in exploring alternative metrics to assess the regression performance for the task.

24 December 2025 in Computer Speech & Language
Verified
Journal Article
Open access
JBJaime Bellver-Soler
Jaime Bellver-Soler
AGAnmol Guragain
Anmol Guragain
SRSamuel Ramos-Varela
Samuel Ramos-Varela
RCRicardo Córdoba
Ricardo Córdoba
LDLuis Fernando D’HAro
Luis Fernando D’HAro

1 May 2026
Verified
Book Chapter
Closed access
SRSamuel Ramos-Varela
Samuel Ramos-Varela
AGAnmol Guragain
Anmol Guragain
JBJaime Bellver-Soler
Jaime Bellver-Soler
DDDavid Aragon Diaz
David Aragon Diaz
LLLong Lin
Long Lin
LDLuis Fernando D'Haro
Luis Fernando D'Haro