ENCP: Episode-Normalized Conformal Prediction for Vision-and-Language Navigation
This research provides a model-agnostic framework for improving safety in VLN agents by better estimating uncertainty during navigation.
This paper introduces ENCP, a method for uncertainty estimation in Vision-Language Navigation (VLN) models.
It addresses the challenge of standard conformal prediction failing on variable-length VLN episodes.ENCP rescales nonconformity scores and calibrates per episode, meeting step-coverage targets on R2R and REVERIE datasets.
Uncertainty estimation for Vision-Language-Navigation (VLN) models is a critical task since it can help identify ambiguous and unreliable predictions, enabling agents to make safer navigation decisions. As one of the most advanced uncertainty estimation frameworks, conformal prediction (CP) offers a promising approach for uncertainty estimation in VLN. However, given that VLN agent requires a sequence of steps, standard calibration in conformal prediction fails to provide coverage guarantee it promises over a dependent, variable-length VLN episode. To this end, we propose Episode-Normalized Conformal Prediction (ENCP), which rescales a nonconformity score by the policy's residual confidence and calibrates one maximum score per episode. Under exchangeable calibration and test episodes, this construction covers the ground truth at every step with probability at least $1 - α$, while allowing dependence among steps within an episode. Across four VLN policies and three nonconformity scores on R2R and REVERIE dataset, ENCP meets all reported empirical step-coverage targets on the seen-to-unseen evaluation. These results demonstrate that ENCP can provide model-agnostic uncertainty estimates, which might be useful for determining when a VLN agent should defer to a more capable predictor, including human assistance.
This report is categorized under AITOP's "Papers" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include ENCP, vision-language navigation, uncertainty estimation, R2R.
Source attribution: curated from arXiv cs.AI (@Vicky Feliren, A. Taufiq Asyhari, Muhamad Risqi U. Saputra).
