A new testing framework for VLN agents combines adaptive test case generation, capability oracles, and feedback to discover more failures and attribute them to specific capability deficiencies more accurately than baselines.
Due to poor planning, the agent did not choose a better route
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.MA 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Where Did It Go Wrong? Capability-Oriented Failure Attribution for Vision-and-Language Navigation Agents
A new testing framework for VLN agents combines adaptive test case generation, capability oracles, and feedback to discover more failures and attribute them to specific capability deficiencies more accurately than baselines.