Salvato in:
Dettagli Bibliografici
Autori principali: Pozos, Victor Sebastian Martinez, Ruiz, Ivan Vladimir Meza
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:https://arxiv.org/abs/2503.04900
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910862408679424
author Pozos, Victor Sebastian Martinez
Ruiz, Ivan Vladimir Meza
author_facet Pozos, Victor Sebastian Martinez
Ruiz, Ivan Vladimir Meza
contents This paper explores the potential of abstracting complex visual information into discrete, structured symbolic sequences using self-supervised learning (SSL). Inspired by how language abstracts and organizes information to enable better reasoning and generalization, we propose a novel approach for generating symbolic representations from visual data. To learn these sequences, we extend the DINO framework to handle visual and symbolic information. Initial experiments suggest that the generated symbolic sequences capture a meaningful level of abstraction, though further refinement is required. An advantage of our method is its interpretability: the sequences are produced by a decoder transformer using cross-attention, allowing attention maps to be linked to specific symbols and offering insight into how these representations correspond to image regions. This approach lays the foundation for creating interpretable symbolic representations with potential applications in high-level scene understanding.
format Preprint
id arxiv_https___arxiv_org_abs_2503_04900
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Extracting Symbolic Sequences from Visual Representations via Self-Supervised Learning
Pozos, Victor Sebastian Martinez
Ruiz, Ivan Vladimir Meza
Computer Vision and Pattern Recognition
Machine Learning
This paper explores the potential of abstracting complex visual information into discrete, structured symbolic sequences using self-supervised learning (SSL). Inspired by how language abstracts and organizes information to enable better reasoning and generalization, we propose a novel approach for generating symbolic representations from visual data. To learn these sequences, we extend the DINO framework to handle visual and symbolic information. Initial experiments suggest that the generated symbolic sequences capture a meaningful level of abstraction, though further refinement is required. An advantage of our method is its interpretability: the sequences are produced by a decoder transformer using cross-attention, allowing attention maps to be linked to specific symbols and offering insight into how these representations correspond to image regions. This approach lays the foundation for creating interpretable symbolic representations with potential applications in high-level scene understanding.
title Extracting Symbolic Sequences from Visual Representations via Self-Supervised Learning
topic Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2503.04900