paper-with-me

Papers

NVILA: Efficient Frontier Visual Language Models

2024-12-05 · CVPR 2025 1 · Zhijian Liu, Ligeng Zhu, Baifeng Shi, Zhuoyang Zhang, Yuming Lou, Shang Yang, Haocheng Xi, Shiyi Cao, Yuxian Gu, Dacheng Li, Xiuyu Li, Yunhao Fang, Yukang Chen, Cheng-Yu Hsieh, De-An Huang, An-Chieh Cheng, Vishwesh Nath, Jinyi Hu, Sifei Liu, Ranjay Krishna, Daguang Xu, Xiaolong Wang, Pavlo Molchanov, Jan Kautz, Hongxu Yin, Song Han, Yao Lu

Visual language models (VLMs) have made significant advances in accuracy in recent years. However, their efficiency has received much less attention. This paper introduces NVILA, a family of open VLMs designed to optimize both efficiency and accuracy. Building on top of VILA, we improve its model architecture by first scaling up the spatial and temporal resolutions, and then compressing visual tokens. This "scale-then-compress" approach enables NVILA to efficiently process high-resolution images and long videos. We also conduct a systematic investigation to enhance the efficiency of NVILA throughout its entire lifecycle, from training and fine-tuning to deployment. NVILA matches or surpasses the accuracy of many leading open and proprietary VLMs across a wide range of image and video benchmarks. At the same time, it reduces training costs by 4.5X, fine-tuning memory usage by 3.4X, pre-filling latency by 1.6-2.2X, and decoding latency by 1.2-2.8X. We will soon make our code and models available to facilitate reproducibility.

📄 PDF Abstract BibTeX arXiv:2412.04468

Code (2)

efficient-large-model/vila pytorch
nvlabs/vila pytorch

Tasks

Video Question Answering

Similar Papers 제목 키워드 기반

AgroOmni: A Large-Scale Multi-view Agricultural Dataset for Cross-Scale Multimodal Reasoning

2026-03-15 · Jiarui Zhang, Junqi Hu, Zurong Mai, Yang Liu 외 arxiv

Modern agricultural data is sourced from diverse platforms and spans multiple spatial scales, ranging from ground-level close-up photography to Unmanned Aerial Vehicle (UAV) aerial observation and satellite remote sensin…

Visual Question AnsweringMultimodal Reasoning

Exploring the Application of Visual Question Answering (VQA) for Classroom Activity Monitoring

2025-07-30 · Sinh Trong Vu, Hieu Trung Pham, Dung Manh Nguyen, Hieu Minh Hoang 외 arxiv

Classroom behavior monitoring is a critical aspect of educational research, with significant implications for student engagement and learning outcomes. Recent advancements in Visual Question Answering (VQA) models offer …

Visual Question Answering

OpenFrontier: General Navigation with Visual-Language Grounded Frontiers

2026-03-05 · Esteban Padilla-Cerdio, Boyang Sun, Marc Pollefeys, Hermann Blum arxiv

Open-world navigation requires robots to make decisions in complex everyday environments while adapting to flexible task requirements. Conventional navigation approaches often rely on dense 3D reconstruction and hand-cra…

Vision-Language Navigation3D Reconstruction

Model-Dowser: Data-Free Importance Probing to Mitigate Catastrophic Forgetting in Multimodal Large Language Models

2026-02-04 · Hyeontaek Hwang, Nguyen Dinh Son, Daeyoung Kim arxiv

Fine-tuning Multimodal Large Language Models (MLLMs) on task-specific data is an effective way to improve performance on downstream applications. However, such adaptation often leads to a degradation in generalization on…

FrontierNet: Learning Visual Cues to Explore

2025-01-08 · Boyang Sun, Hanzhi Chen, Stefan Leutenegger, Cesar Cadena 외

Exploration of unknown environments is crucial for autonomous robots; it allows them to actively reason and decide on what new data to acquire for different tasks, such as mapping, object discovery, and environmental ass…

Object Discovery