Intelligent support for Human Oversight: Integrating Reinforcement Learning with Gaze Simulation to Personalize Highlighting
Not on the map. Why?
Abstract (arXiv)
Interfaces for human oversight must effectively support users' situation awareness under time-critical conditions. We explore reinforcement learning (RL)-based UI adaptation to personalize alerting strategies that balance the benefits of highlighting critical events against the cognitive costs of interruptions. To enable learning without real-world deployment, we integrate models of users' gaze behavior to simulate attentional dynamics during monitoring. Using a delivery-drone oversight scenario, we present initial results suggesting that RL-based highlighting can outperform static, rule-based approaches and discuss challenges of intelligent oversight support.
Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 12 of its 20 references that are in the map. First posted on arXiv 9 Feb 2026.
Builds on
12 of 20 references.
References · 12
- 2017Proximal Policy Optimization Algorithmscited by 15,215
- 2020What Matters In On-Policy Reinforcement Learning? A Large-Scale Empirical Studycited by 162
- 2019TASED-Net: Temporally-Aggregating Spatial Encoder-Decoder Network for Video Saliency Detectioncited by 59
- 2024On the Quest for Effectiveness in Human Oversight: Interdisciplinary Perspectivescited by 28
- 2023TempSAL -- Uncovering Temporal Information for Deep Saliency Predictioncited by 25
- 2024EyeFormer: Predicting Personalized Scanpaths with Transformer-Guided Reinforcement Learningcited by 8
- 2024Shifting Focus with HCEye: Exploring the Dynamics of Visual Highlighting and Cognitive Load on User Attention and Saliency Predictioncited by 6
- 2020Detecting Relevance during Decision-Making from Eye Movements for UI Adaptationcited by 5
- 2024Reinforcement Learning-Based Framework for the Intelligent Adaptation of User Interfacescited by 3
- 2024Enhancing Saliency Prediction in Monitoring Tasks: The Role of Visual Highlightscited by 1
- 2024The Design of Informative Take-Over Requests for Semi-Autonomous Cyber-Physical Systems: Combining Spoken Language and Visual Icons in a Drone-Controller Settingcited by 1
- 2025Understanding and Predicting Temporal Visual Attention Influenced by Dynamic Highlights in Monitoring Taskcited by 1
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2025Enhancing RLHF with Human Gaze Modeling
- 2025Assessing Policy Updates: Toward Trust-Preserving Intelligent User Interfaces
- 2025Integrating Human Feedback into a Reinforcement Learning-Based Framework for Adaptive User Interfaces
- 2026Steering LLMs via Scalable Interactive Oversight
- 2026From Gaze to Guidance: Interpreting and Adapting to Users' Cognitive Needs with Multimodal Gaze-Aware AI Assistants
- 2025Enhancing Visual Grounding for GUI Agents via Self-Evolutionary Reinforcement Learning
- 2025Revealing Human Attention Patterns from Gameplay Analysis for Reinforcement Learning
- 2026Anticipatory Human Oversight of Agentic AI: A Philosophical Account