Temporal Difference Calibration in Sequential Tasks: Application to Vision-Language-Action Models
This addresses the challenge of reliable uncertainty quantification in episodic robotics tasks, which is incremental but important for improving model reliability.
The paper tackles the problem of uncertainty calibration in vision-language-action models for sequential robotics tasks, showing that temporal-difference calibration improves performance on simulated and real-robot data, with competitive uncertainty estimates from single-step action probabilities.
Recent advances in vision-language-action (VLA) models for robotics have highlighted the importance of reliable uncertainty quantification in sequential tasks. However, assessing and improving calibration in such settings remains mostly unexplored, especially when only partial trajectories are observed. In this work, we formulate sequential calibration for episodic tasks, where task-success confidence is produced along an episode, while success is determined at the end of it. We introduce a sequential extension of the Brier score and show that, for binary outcomes, its risk minimizer coincides with the VLA policy's value function. This connection bridges uncertainty calibration and reinforcement learning, enabling the use of temporal-difference (TD) value estimation as a principled calibration mechanism over time. We empirically show that TD calibration improves performance relative to the state-of-the-art on simulated and real-robot data. Interestingly, we show that when calibrated using TD, the VLA's single-step action probabilities can yield competitive uncertainty estimates, in contrast to recent findings that employed different calibration techniques.