@article{Johannes A. Gaus2026,
author = "Johannes A. Gaus and  Winfried Ilg and Daniel Haeufle",
abstract = "Assistive devices must determine both what a
user intends to do and how reliable that prediction is before
providing support. We introduce a safety–critical triggering
framework based on calibrated probabilities for multimodal
next–action prediction in Activities of Daily Living. Raw model
confidence often fails to reflect true correctness, posing a safety
risk. Post–hoc calibration aligns predicted confidence with
empirical reliability and reduces miscalibration by about an
order of magnitude without affecting accuracy. The calibrated
confidence drives a simple ACT/HOLD rule that acts only when
reliability is high and withholds assistance otherwise. This turns
the confidence threshold into a quantitative safety parameter for
assisted actions and enables verifiable behavior in an assistive
control loop",
doi = " https://doi.org/10.48550/arXiv.2601.04982",
journal = "arXiv",
title = "{W}hen to {A}ct: {C}alibrated {C}onfidence for {R}eliable {H}uman {I}ntention {P}rediction in {A}ssistive {R}obotics",
url = "https://arxiv.org/pdf/2601.04982",
year = "2026",
files = "2601.04982v1.pdf",
}