Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Bingyang, Li, Yijiang, Qiao, Yitong, Wang, Maijunxian, Zhao, Tianwei, Sun, Yucheng, Deng, Binyue, Deng, Hokin, Vasconcelos, Nuno, Luo, Dezhi
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2505.18969
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866912930453258240
author Wang, Bingyang
Li, Yijiang
Qiao, Yitong
Wang, Maijunxian
Zhao, Tianwei
Sun, Yucheng
Deng, Binyue
Deng, Hokin
Vasconcelos, Nuno
Luo, Dezhi
author_facet Wang, Bingyang
Li, Yijiang
Qiao, Yitong
Wang, Maijunxian
Zhao, Tianwei
Sun, Yucheng
Deng, Binyue
Deng, Hokin
Vasconcelos, Nuno
Luo, Dezhi
contents Cognitive control, the ability to coordinate competing information sources in pursuit of goals, is fundamental to intelligent behavior. We systematically investigate whether Vision Language Models (VLMs) exhibit cognitive control and how computational resources modulate conflict resolution. We construct a benchmark of 4,410 tasks across seven conflict paradigms (Stroop, Flanker, and five realistic variants) spanning multiple difficulty levels and visual complexities, testing 47 VLMs with rigorous experimental control. We find that VLMs exhibit robust congruency effects across all tasks, with larger models systematically resolving conflicts more effectively than smaller models. Critically, VLMs reproduce the fine-grained demand-resource relationship observed in human temporal dynamics: larger models drop below chance on incongruent high-conflict trials while smaller models fail to meaningfully engage and perform at chance, mirroring human behavior at short processing times and establishing parameter count as a proxy for conflict resolution capacity. These findings demonstrate that human-like cognitive control emerges from optimization dynamics in large-scale neural networks, suggesting that adaptive flexibility under conflict may naturally arise through scaling.
format Preprint
id arxiv_https___arxiv_org_abs_2505_18969
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Increasing Computation Resolves Conflicts in Vision Language Models
Wang, Bingyang
Li, Yijiang
Qiao, Yitong
Wang, Maijunxian
Zhao, Tianwei
Sun, Yucheng
Deng, Binyue
Deng, Hokin
Vasconcelos, Nuno
Luo, Dezhi
Neural and Evolutionary Computing
Cognitive control, the ability to coordinate competing information sources in pursuit of goals, is fundamental to intelligent behavior. We systematically investigate whether Vision Language Models (VLMs) exhibit cognitive control and how computational resources modulate conflict resolution. We construct a benchmark of 4,410 tasks across seven conflict paradigms (Stroop, Flanker, and five realistic variants) spanning multiple difficulty levels and visual complexities, testing 47 VLMs with rigorous experimental control. We find that VLMs exhibit robust congruency effects across all tasks, with larger models systematically resolving conflicts more effectively than smaller models. Critically, VLMs reproduce the fine-grained demand-resource relationship observed in human temporal dynamics: larger models drop below chance on incongruent high-conflict trials while smaller models fail to meaningfully engage and perform at chance, mirroring human behavior at short processing times and establishing parameter count as a proxy for conflict resolution capacity. These findings demonstrate that human-like cognitive control emerges from optimization dynamics in large-scale neural networks, suggesting that adaptive flexibility under conflict may naturally arise through scaling.
title Increasing Computation Resolves Conflicts in Vision Language Models
topic Neural and Evolutionary Computing
url https://arxiv.org/abs/2505.18969