BARE: Towards Bias-Aware and Reasoning-Enhanced One-Tower Visual Grounding
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2026
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866909981145563136 |
|---|---|
| author | Li, Hongbing Xiao, Linhui Zhao, Zihan Shen, Qi Huang, Yixiang Xiao, Bo Ma, Zhanyu |
| author_facet | Li, Hongbing Xiao, Linhui Zhao, Zihan Shen, Qi Huang, Yixiang Xiao, Bo Ma, Zhanyu |
| contents | Visual Grounding (VG), which aims to locate a specific region referred to by expressions, is a fundamental yet challenging task in the multimodal understanding fields. While recent grounding transfer works have advanced the field through one-tower architectures, they still suffer from two primary limitations: (1) over-entangled multimodal representations that exacerbate deceptive modality biases, and (2) insufficient semantic reasoning that hinders the comprehension of referential cues. In this paper, we propose BARE, a bias-aware and reasoning-enhanced framework for one-tower visual grounding. BARE introduces a mechanism that preserves modality-specific features and constructs referential semantics through three novel modules: (i) language salience modulator, (ii) visual bias correction and (iii) referential relationship enhancement, which jointly mitigate multimodal distractions and enhance referential comprehension. Extensive experimental results on five benchmarks demonstrate that BARE not only achieves state-of-the-art performance but also delivers superior computational efficiency compared to existing approaches. The code is publicly accessible at https://github.com/Marloweeee/BARE. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_01526 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | BARE: Towards Bias-Aware and Reasoning-Enhanced One-Tower Visual Grounding Li, Hongbing Xiao, Linhui Zhao, Zihan Shen, Qi Huang, Yixiang Xiao, Bo Ma, Zhanyu Computer Vision and Pattern Recognition Visual Grounding (VG), which aims to locate a specific region referred to by expressions, is a fundamental yet challenging task in the multimodal understanding fields. While recent grounding transfer works have advanced the field through one-tower architectures, they still suffer from two primary limitations: (1) over-entangled multimodal representations that exacerbate deceptive modality biases, and (2) insufficient semantic reasoning that hinders the comprehension of referential cues. In this paper, we propose BARE, a bias-aware and reasoning-enhanced framework for one-tower visual grounding. BARE introduces a mechanism that preserves modality-specific features and constructs referential semantics through three novel modules: (i) language salience modulator, (ii) visual bias correction and (iii) referential relationship enhancement, which jointly mitigate multimodal distractions and enhance referential comprehension. Extensive experimental results on five benchmarks demonstrate that BARE not only achieves state-of-the-art performance but also delivers superior computational efficiency compared to existing approaches. The code is publicly accessible at https://github.com/Marloweeee/BARE. |
| title | BARE: Towards Bias-Aware and Reasoning-Enhanced One-Tower Visual Grounding |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2601.01526 |