Where Did It Go Wrong? Capability-Oriented Failure Attribution for Vision-and-Language Navigation Agents

TL;DR

This paper introduces a capability-oriented testing method for Vision-and-Language Navigation agents that improves failure detection, attribution, and interpretability over existing system-level approaches.

Contribution

It presents a novel adaptive testing framework combining seed mutation, capability oracles, and feedback for precise failure attribution in embodied agents.

Findings

Discoveries include more failure cases than baseline methods.

Accurately pinpoints capability-specific deficiencies.

Provides actionable insights for agent improvement.

Abstract

Embodied agents in safety-critical applications such as Vision-Language Navigation (VLN) rely on multiple interdependent capabilities (e.g., perception, memory, planning, decision), making failures difficult to localize and attribute. Existing testing methods are largely system-level and provide limited insight into which capability deficiencies cause task failures. We propose a capability-oriented testing approach that enables failure detection and attribution by combining (1) adaptive test case generation via seed selection and mutation, (2) capability oracles for identifying capability-specific errors, and (3) a feedback mechanism that attributes failures to capabilities and guides further test generation. Experiments show that our method discovers more failure cases and more accurately pinpoints capability-level deficiencies than state-of-the-art baselines, providing more interpretable and actionable guidance for improving embodied agents.