DriveCritic: Towards Context-Aware, Human-Aligned Evaluation for Autonomous Driving with Vision-Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Song, Jingyu, Li, Zhenxin, Lan, Shiyi, Sun, Xinglong, Chang, Nadine, Shen, Maying, Chen, Joshua, Skinner, Katherine A., Alvarez, Jose M.
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908880878960640
author Song, Jingyu
Li, Zhenxin
Lan, Shiyi
Sun, Xinglong
Chang, Nadine
Shen, Maying
Chen, Joshua
Skinner, Katherine A.
Alvarez, Jose M.
author_facet Song, Jingyu
Li, Zhenxin
Lan, Shiyi
Sun, Xinglong
Chang, Nadine
Shen, Maying
Chen, Joshua
Skinner, Katherine A.
Alvarez, Jose M.
contents Benchmarking autonomous driving planners to align with human judgment remains a critical challenge, as state-of-the-art metrics like the Extended Predictive Driver Model Score (EPDMS) lack context awareness in nuanced scenarios. To address this, we introduce DriveCritic, a novel framework featuring two key contributions: the DriveCritic dataset, a curated collection of challenging scenarios where context is critical for correct judgment and annotated with pairwise human preferences, and the DriveCritic model, a Vision-Language Model (VLM) based evaluator. Fine-tuned using a two-stage supervised and reinforcement learning pipeline, the DriveCritic model learns to adjudicate between trajectory pairs by integrating visual and symbolic context. Experiments show DriveCritic significantly outperforms existing metrics and baselines in matching human preferences and demonstrates strong context awareness. Overall, our work provides a more reliable, human-aligned foundation to evaluating autonomous driving systems. The project page for DriveCritic is https://song-jingyu.github.io/DriveCritic
format Preprint
id arxiv_https___arxiv_org_abs_2510_13108
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle DriveCritic: Towards Context-Aware, Human-Aligned Evaluation for Autonomous Driving with Vision-Language Models
Song, Jingyu
Li, Zhenxin
Lan, Shiyi
Sun, Xinglong
Chang, Nadine
Shen, Maying
Chen, Joshua
Skinner, Katherine A.
Alvarez, Jose M.
Computer Vision and Pattern Recognition
Artificial Intelligence
Robotics
Benchmarking autonomous driving planners to align with human judgment remains a critical challenge, as state-of-the-art metrics like the Extended Predictive Driver Model Score (EPDMS) lack context awareness in nuanced scenarios. To address this, we introduce DriveCritic, a novel framework featuring two key contributions: the DriveCritic dataset, a curated collection of challenging scenarios where context is critical for correct judgment and annotated with pairwise human preferences, and the DriveCritic model, a Vision-Language Model (VLM) based evaluator. Fine-tuned using a two-stage supervised and reinforcement learning pipeline, the DriveCritic model learns to adjudicate between trajectory pairs by integrating visual and symbolic context. Experiments show DriveCritic significantly outperforms existing metrics and baselines in matching human preferences and demonstrates strong context awareness. Overall, our work provides a more reliable, human-aligned foundation to evaluating autonomous driving systems. The project page for DriveCritic is https://song-jingyu.github.io/DriveCritic
title DriveCritic: Towards Context-Aware, Human-Aligned Evaluation for Autonomous Driving with Vision-Language Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Robotics
url https://arxiv.org/abs/2510.13108