In a simulated robot teleoperation task, choosing the inference with higher confidence between human and AI improves team accuracy when the AI's confidence is well calibrated, and hurts when it is poorly calibrated.
In Conference on Human Factors in Computing Systems (CHI)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.HC 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Enhancing Joint Human-AI Inference in Robot Missions: A Confidence-Based Approach
In a simulated robot teleoperation task, choosing the inference with higher confidence between human and AI improves team accuracy when the AI's confidence is well calibrated, and hurts when it is poorly calibrated.