Saved in:
Bibliographic Details
Main Authors: Yildiz, Anil, Thornton, Sarah M., Hildebrandt, Carl, Roy-Singh, Sreeja, Kochenderfer, Mykel J.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2510.24949
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917049116131328
author Yildiz, Anil
Thornton, Sarah M.
Hildebrandt, Carl
Roy-Singh, Sreeja
Kochenderfer, Mykel J.
author_facet Yildiz, Anil
Thornton, Sarah M.
Hildebrandt, Carl
Roy-Singh, Sreeja
Kochenderfer, Mykel J.
contents Assessing scenario coverage is crucial for evaluating the robustness of autonomous agents, yet existing methods rely on expensive human annotations or computationally intensive Large Vision-Language Models (LVLMs). These approaches are impractical for large-scale deployment due to cost and efficiency constraints. To address these shortcomings, we propose SCOUT (Scenario Coverage Oversight and Understanding Tool), a lightweight surrogate model designed to predict scenario coverage labels directly from an agent's latent sensor representations. SCOUT is trained through a distillation process, learning to approximate LVLM-generated coverage labels while eliminating the need for continuous LVLM inference or human annotation. By leveraging precomputed perception features, SCOUT avoids redundant computations and enables fast, scalable scenario coverage estimation. We evaluate our method across a large dataset of real-life autonomous navigation scenarios, demonstrating that it maintains high accuracy while significantly reducing computational cost. Our results show that SCOUT provides an effective and practical alternative for large-scale coverage analysis. While its performance depends on the quality of LVLM-generated training labels, SCOUT represents a major step toward efficient scenario coverage oversight in autonomous systems.
format Preprint
id arxiv_https___arxiv_org_abs_2510_24949
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SCOUT: A Lightweight Framework for Scenario Coverage Assessment in Autonomous Driving
Yildiz, Anil
Thornton, Sarah M.
Hildebrandt, Carl
Roy-Singh, Sreeja
Kochenderfer, Mykel J.
Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
Assessing scenario coverage is crucial for evaluating the robustness of autonomous agents, yet existing methods rely on expensive human annotations or computationally intensive Large Vision-Language Models (LVLMs). These approaches are impractical for large-scale deployment due to cost and efficiency constraints. To address these shortcomings, we propose SCOUT (Scenario Coverage Oversight and Understanding Tool), a lightweight surrogate model designed to predict scenario coverage labels directly from an agent's latent sensor representations. SCOUT is trained through a distillation process, learning to approximate LVLM-generated coverage labels while eliminating the need for continuous LVLM inference or human annotation. By leveraging precomputed perception features, SCOUT avoids redundant computations and enables fast, scalable scenario coverage estimation. We evaluate our method across a large dataset of real-life autonomous navigation scenarios, demonstrating that it maintains high accuracy while significantly reducing computational cost. Our results show that SCOUT provides an effective and practical alternative for large-scale coverage analysis. While its performance depends on the quality of LVLM-generated training labels, SCOUT represents a major step toward efficient scenario coverage oversight in autonomous systems.
title SCOUT: A Lightweight Framework for Scenario Coverage Assessment in Autonomous Driving
topic Robotics
Artificial Intelligence
Computer Vision and Pattern Recognition
Machine Learning
url https://arxiv.org/abs/2510.24949