Saved in:
Bibliographic Details
Main Authors: Wang, Chaoqun, Yang, Jie, Hong, Xiaobin, Zhang, Ruimao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2503.10586
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917958002933760
author Wang, Chaoqun
Yang, Jie
Hong, Xiaobin
Zhang, Ruimao
author_facet Wang, Chaoqun
Yang, Jie
Hong, Xiaobin
Zhang, Ruimao
contents Recent Vision-based Large Language Models~(VisionLLMs) for autonomous driving have seen rapid advancements. However, such promotion is extremely dependent on large-scale high-quality annotated data, which is costly and labor-intensive. To address this issue, we propose unlocking the value of abundant yet unlabeled data to improve the language-driving model in a semi-supervised learning manner. Specifically, we first introduce a series of template-based prompts to extract scene information, generating questions that create pseudo-answers for the unlabeled data based on a model trained with limited labeled data. Next, we propose a Self-Consistency Refinement method to improve the quality of these pseudo-annotations, which are later used for further training. By utilizing a pre-trained VisionLLM (e.g., InternVL), we build a strong Language Driving Model (LDM) for driving scene question-answering, outperforming previous state-of-the-art methods. Extensive experiments on the DriveLM benchmark show that our approach performs well with just 5% labeled data, achieving competitive performance against models trained with full datasets. In particular, our LDM achieves 44.85% performance with limited labeled data, increasing to 54.27% when using unlabeled data, while models trained with full datasets reach 60.68% on the DriveLM benchmark.
format Preprint
id arxiv_https___arxiv_org_abs_2503_10586
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Unlock the Power of Unlabeled Data in Language Driving Model
Wang, Chaoqun
Yang, Jie
Hong, Xiaobin
Zhang, Ruimao
Computer Vision and Pattern Recognition
Recent Vision-based Large Language Models~(VisionLLMs) for autonomous driving have seen rapid advancements. However, such promotion is extremely dependent on large-scale high-quality annotated data, which is costly and labor-intensive. To address this issue, we propose unlocking the value of abundant yet unlabeled data to improve the language-driving model in a semi-supervised learning manner. Specifically, we first introduce a series of template-based prompts to extract scene information, generating questions that create pseudo-answers for the unlabeled data based on a model trained with limited labeled data. Next, we propose a Self-Consistency Refinement method to improve the quality of these pseudo-annotations, which are later used for further training. By utilizing a pre-trained VisionLLM (e.g., InternVL), we build a strong Language Driving Model (LDM) for driving scene question-answering, outperforming previous state-of-the-art methods. Extensive experiments on the DriveLM benchmark show that our approach performs well with just 5% labeled data, achieving competitive performance against models trained with full datasets. In particular, our LDM achieves 44.85% performance with limited labeled data, increasing to 54.27% when using unlabeled data, while models trained with full datasets reach 60.68% on the DriveLM benchmark.
title Unlock the Power of Unlabeled Data in Language Driving Model
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2503.10586