Research map / Large language models / AI and society
Neuro-Cognitive Reward Modeling for Human-Centered Autonomous Vehicle Control
Zhuoli Zhuang, Yu-Cheng Chang, Yu-Kai Wang, Thomas Do, Chin-Teng Lin
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| Add EEG-predicted cognitive reward to TD3 reward | TD3 (vanilla) · Emergency Braking | Driving Score: 16 -> 79 | Improved |
| Add EEG-predicted cognitive reward to TD3 reward | TD3 (vanilla) · Left Turn | Driving Score: 45 -> 57 | Improved |
| Use EEG-based cognitive reward instead of human-preference RLHF reward model | TD3 · Emergency Braking | Driving Score: 66 -> 79 | Improved |
| Use EEG-based cognitive reward instead of human-preference RLHF reward model | TD3 · Left Turn | Driving Score: 49 -> 57 | No clear effect |
| Initialize policy with behavior cloning | TD3 · Left Turn | Driving Score: 45 -> 29 | Hurt |
| Add Lagrangian safety constraint (TD3-lag) | TD3 · Emergency Braking | Driving Score: 16 -> 35 | Improved |
| Prioritized human-in-the-loop replay (PHIL) instead of vanilla TD3 | TD3 · Left Turn | Driving Score: 45 -> 31 | Hurt |
| Use lightweight 3-layer CNN instead of ResNet-18 for ERP prediction | EEG feature prediction model · ERP prediction, five-fold CV | mean accuracy (%): 81 -> 82 | No clear effect |
| Use lightweight 3-layer CNN instead of ResNet-18 for ERP prediction | EEG feature prediction model · ERP prediction | FPS: 73 -> 204 | Improved |
| Human responds actively to braking vs not required to react | ERP amplitude · Emergency braking EEG, 20 participants | ERP amplitude 300-500 ms | Improved |
Closest papers by meaning
- Embodied Cognition Augmented End2End Autonomous Driving (NeurIPS 2025 · similarity 0.87)
- Distill to Think, Foresee to Act: Cognitive-Physical Reinforcement Learning for Autonomous Driving (NeurIPS 2026 · similarity 0.86)
- Continuously Learning, Adapting, and Improving: A Dual-Process Approach to Autonomous Driving (NeurIPS 2024 · similarity 0.85)
- Prioritizing Perception-Guided Self-Supervision: A New Paradigm for Causal Modeling in End-to-End Autonomous Driving (NeurIPS 2025 · similarity 0.84)
- Estimating cognitive biases with attention-aware inverse planning (NeurIPS 2025 · similarity 0.84)
- DriveAgent-R1: Advancing VLM-based Autonomous Driving with Active Perception and Hybrid Thinking (ICLR 2026 · similarity 0.83)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
