mlsanityData to 13 Oct 2026

When to Act: Calibrated Confidence for Reliable Human Intention Prediction in Assistive Robotics

arXiv 2601.04982 ↗First posted 8 Jan 2026cs.RO cs.AIcited by 0

Not on the map. Why?

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 13 Oct 2026). Builds on 17 of its 25 references that are in the map. First posted on arXiv 8 Jan 2026.

Builds on

17 of 25 references.

References · 17

  1. 2017On Calibration of Modern Neural Networkscited by 4,902
  2. 2015Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learningcited by 4,430
  3. 2016Simple and Scalable Predictive Uncertainty Estimation using Deep Ensemblescited by 3,902
  4. 2017What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?cited by 2,340
  5. 2017Selective Classification for Deep Neural Networkscited by 825
  6. 2021A Survey of Uncertainty in Deep Neural Networkscited by 676
  7. 2020Rescaling Egocentric Visioncited by 532
  8. 2021Laplace Redux -- Effortless Bayesian Deep Learningcited by 361
  9. 2019SelectiveNet: A Deep Neural Network with an Integrated Reject Optioncited by 341
  10. 2021Anticipative Video Transformercited by 199
  11. 2019What Would You Expect? Anticipating Egocentric Actions with Rolling-Unrolling LSTMs and Modality Attentioncited by 105
  12. 2020In the Eye of the Beholder: Gaze and Actions in First Person Videocited by 69
5 more
  1. 2017Shared Autonomy via Hindsight Optimization for Teleoperation and Teamingcited by 39
  2. 2019Uncertainty-Aware Anticipation of Activitiescited by 39
  3. 2024HOIMotion: Forecasting Human Motion During Human-Object Interactions Using Egocentric 3D Object Bounding Boxescited by 16
  4. 2023A Survey on Deep Learning Techniques for Action Anticipationcited by 11
  5. 2023EPIC-KITCHENS-100 Unsupervised Domain Adaptation Challenge: Mixed Sequences Predictioncited by 3

Built on by

Not cited yet.