VISOR: A Vision-Language Model-based Test Oracle for Testing Robots

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Saurabh, Prasun, Valle, Pablo, Arrieta, Aitor, Ali, Shaukat, Arcaini, Paolo
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916017540694016
author Saurabh, Prasun
Valle, Pablo
Arrieta, Aitor
Ali, Shaukat
Arcaini, Paolo
author_facet Saurabh, Prasun
Valle, Pablo
Arrieta, Aitor
Ali, Shaukat
Arcaini, Paolo
contents Testing robots requires assessing whether they perform their intended tasks correctly, dependably, and with high quality, a challenge known as the test oracle problem in software testing. Traditionally, this assessment relies on task-specific symbolic oracles for task correctness and on human manual evaluation of robot behavior, which is time-consuming, subjective, and error-prone. To address this, we propose VISOR, a Vision-Language Model (VLM)-based approach for automated test oracle assessment that eliminates the need of expensive human evaluations. VISOR performs automated evaluation of task correctness and quality, addressing the limitations of existing symbolic test oracles, which are task-specific and provide pass/fail judgments without explicitly quantifying task quality. Given the inherent uncertainty in VLMs, VISOR also explicitly quantifies its own uncertainty during test assessments. We evaluated VISOR using two VLMs, i.e., GPT and Gemini, across four robotic tasks on over 1,000 videos. Results show that Gemini achieves higher recall while GPT achieves higher precision. However, both models show low correlation between uncertainty and correctness, which prevents using uncertainty as a correctness predictor.
format Preprint
id arxiv_https___arxiv_org_abs_2605_10408
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle VISOR: A Vision-Language Model-based Test Oracle for Testing Robots
Saurabh, Prasun
Valle, Pablo
Arrieta, Aitor
Ali, Shaukat
Arcaini, Paolo
Software Engineering
Robotics
Testing robots requires assessing whether they perform their intended tasks correctly, dependably, and with high quality, a challenge known as the test oracle problem in software testing. Traditionally, this assessment relies on task-specific symbolic oracles for task correctness and on human manual evaluation of robot behavior, which is time-consuming, subjective, and error-prone. To address this, we propose VISOR, a Vision-Language Model (VLM)-based approach for automated test oracle assessment that eliminates the need of expensive human evaluations. VISOR performs automated evaluation of task correctness and quality, addressing the limitations of existing symbolic test oracles, which are task-specific and provide pass/fail judgments without explicitly quantifying task quality. Given the inherent uncertainty in VLMs, VISOR also explicitly quantifies its own uncertainty during test assessments. We evaluated VISOR using two VLMs, i.e., GPT and Gemini, across four robotic tasks on over 1,000 videos. Results show that Gemini achieves higher recall while GPT achieves higher precision. However, both models show low correlation between uncertainty and correctness, which prevents using uncertainty as a correctness predictor.
title VISOR: A Vision-Language Model-based Test Oracle for Testing Robots
topic Software Engineering
Robotics
url https://arxiv.org/abs/2605.10408