Viper-F1: Fast and Fine-Grained Multimodal Understanding with Cross-Modal State-Space Modulation

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
1. Verfasser: Trinh, Quoc-Huy
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914170722582528
author Trinh, Quoc-Huy
author_facet Trinh, Quoc-Huy
contents Recent advances in multimodal large language models (MLLMs) have enabled impressive progress in vision-language understanding, yet their high computational cost limits deployment in resource-constrained scenarios such as robotic manipulation, personal assistants, and smart cameras. Most existing methods rely on Transformer-based cross-attention, whose quadratic complexity hinders efficiency. Moreover, small vision-language models often struggle to precisely capture fine-grained, task-relevant visual regions, leading to degraded performance on fine-grained reasoning tasks that limit their effectiveness in the real world. To address these issues, we introduce Viper-F1, a Hybrid State-Space Vision-Language Model that replaces attention with efficient Liquid State-Space Dynamics. To further enhance visual grounding, we propose a Token-Grid Correlation Module, which computes lightweight correlations between text tokens and image patches and modulates the state-space dynamics via FiLM conditioning. This enables the model to selectively emphasize visual regions relevant to the textual prompt while maintaining linear-time inference. Experimental results across multiple benchmarks demonstrate that Viper-F1 achieves accurate, fine-grained understanding with significantly improved efficiency.
format Preprint
id arxiv_https___arxiv_org_abs_2511_11177
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Viper-F1: Fast and Fine-Grained Multimodal Understanding with Cross-Modal State-Space Modulation
Trinh, Quoc-Huy
Computer Vision and Pattern Recognition
Recent advances in multimodal large language models (MLLMs) have enabled impressive progress in vision-language understanding, yet their high computational cost limits deployment in resource-constrained scenarios such as robotic manipulation, personal assistants, and smart cameras. Most existing methods rely on Transformer-based cross-attention, whose quadratic complexity hinders efficiency. Moreover, small vision-language models often struggle to precisely capture fine-grained, task-relevant visual regions, leading to degraded performance on fine-grained reasoning tasks that limit their effectiveness in the real world. To address these issues, we introduce Viper-F1, a Hybrid State-Space Vision-Language Model that replaces attention with efficient Liquid State-Space Dynamics. To further enhance visual grounding, we propose a Token-Grid Correlation Module, which computes lightweight correlations between text tokens and image patches and modulates the state-space dynamics via FiLM conditioning. This enables the model to selectively emphasize visual regions relevant to the textual prompt while maintaining linear-time inference. Experimental results across multiple benchmarks demonstrate that Viper-F1 achieves accurate, fine-grained understanding with significantly improved efficiency.
title Viper-F1: Fast and Fine-Grained Multimodal Understanding with Cross-Modal State-Space Modulation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.11177