Do What? Teaching Vision-Language-Action Models to Reject the Impossible
This addresses a critical issue for robotic systems using VLA models by enabling them to reject impossible commands, though it is an incremental advancement in robustness.
The paper tackles the problem of Vision-Language-Action (VLA) models handling false-premise instructions, proposing the Instruct-Verify-and-Act (IVA) framework to detect, clarify, and correct such requests, resulting in a 97.56% improvement in detection accuracy and a 50.78% increase in successful responses.
Recently, Vision-Language-Action (VLA) models have demonstrated strong performance on a range of robotic tasks. These models rely on multimodal inputs, with language instructions playing a crucial role -- not only in predicting actions, but also in robustly interpreting user intent, even when the requests are impossible to fulfill. In this work, we investigate how VLAs can recognize, interpret, and respond to false-premise instructions: natural language commands that reference objects or conditions absent from the environment. We propose Instruct-Verify-and-Act (IVA), a unified framework that (i) detects when an instruction cannot be executed due to a false premise, (ii) engages in language-based clarification or correction, and (iii) grounds plausible alternatives in perception and action. Towards this end, we construct a large-scale instruction tuning setup with structured language prompts and train a VLA model capable of handling both accurate and erroneous requests. Our approach leverages a contextually augmented, semi-synthetic dataset containing paired positive and false-premise instructions, enabling robust detection and natural language correction. Our experiments show that IVA improves false premise detection accuracy by 97.56% over baselines, while increasing successful responses in false-premise scenarios by 50.78%.