Research map / Vision-language and multimodal / Human and embodied modeling
Grounding Human-To-Vehicle Advice for Self-Driving Vehicles
Check your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 natural-language advice to the controller | End-to-end driving controller · HAD test | Steering median absolute error: 3.92 [1.54, 10.1] -> 3.53 [1.37, 8.83] | Helped |
| Use stimulus-driven advice | End-to-end driving controller · HAD test | Speed median absolute error: 3.98 [1.76, 8.10] -> 3.28 [1.47, 6.46] | Helped |
| Use stimulus-driven advice instead of goal-oriented advice | End-to-end driving controller · HAD test | Speed median absolute error: 4.25 [1.86, 8.46] -> 3.28 [1.47, 6.46] | Helped |
| Use goal-oriented advice instead of stimulus-driven advice | End-to-end driving controller · HAD test | Steering median absolute error: 3.78 [1.45, 9.93] -> 3.53 [1.37, 8.83] | Helped |
| Train on both advice types and test with either type | End-to-end driving controller · HAD test | Speed median absolute error: 3.78 [1.67, 7.50]; 3.78 [1.68, 7.46] | Mixed |
| Add attention-map internalization loss (ablation) | Advice-taking controller when advice is unavailable at test · HAD test | Speed median absolute error: 3.55 [1.58, 7.12] -> 3.36 [1.51, 6.62] | Helped |
| Add attention-map internalization loss (ablation) | Advice-taking controller when advice is unavailable at test · HAD test | Steering median absolute error: 4.01 [1.59, 10.1] -> 3.96 [1.55, 10.0] | Helped |
| Replace base visual encoder with Inception v3 (ablation) | Advice-taking controller · HAD test | Speed median absolute error: 3.78 [1.68, 7.49] -> 2.89 [1.31, 5.59] | Helped |
| Replace visual encoder with MobileNet (ablation) | Advice-taking controller · HAD test | Steering median absolute error: 3.58 [1.39, 9.34] -> 4.20 [1.65, 10.7] | Hurt |
| Add integral loss term (ablation) | Advice-taking controller · HAD test | Speed median absolute error | Helped |
| Add derivative loss term (ablation) | Advice-taking controller · HAD test | Steering median absolute error | Hurt |
Closest papers by meaning
- Less Language, More Latents: Annotation-Efficient VLAs for Driving (NeurIPS 2026 · similarity 0.85)
- Prioritizing Perception-Guided Self-Supervision: A New Paradigm for Causal Modeling in End-to-End Autonomous Driving (NeurIPS 2025 · similarity 0.85)
- Feedback-Guided Autonomous Driving (CVPR 2024 · similarity 0.85)
- Language Prompt for Autonomous Driving (AAAI 2025 · similarity 0.83)
- Continuously Learning, Adapting, and Improving: A Dual-Process Approach to Autonomous Driving (NeurIPS 2024 · similarity 0.83)
- Enhancing End-to-End Autonomous Driving with Latent World Model (ICLR 2025 · 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.