Paper Overview
Field: Machine Learning Authors: Vicky Feliren, A. Taufiq Asyhari, Muhamad Risqi U. Saputra Published: 2026-09-15 arXiv: 2609.17499
Summary
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 a VLN agent requires a sequence of steps, standard calibration in conformal prediction fails to provide the coverage guarantee it promises over a dependent, variable-length VLN episode.
To address this, the authors propose Episode-Normalized Conformal Prediction (ENCP), which:
- Rescales a nonconformity score by the policy's residual confidence
- Calibrates one maximum score per episode
- arXiv page: https://arxiv.org/abs/2609.17499
Under exchangeable calibration and test episodes, this construction covers the ground truth at each step with probability at least 1-α, while allowing dependencies between steps within an episode.
Results
On the R2R and REVERIE datasets, across 4 VLN policies and 3 nonconformity scores, ENCP achieves all reported empirical per-step coverage targets in both seen and unseen evaluations. These results show that ENCP can provide model-agnostic uncertainty estimation, which may help determine when a VLN agent should switch to a stronger predictor, including human assistance.