Research map / Large language models / Agents, code and math
InAction: Interpretable Action Decision Making for Autonomous Driving
Taotao Jing, Haifeng Xia, Renran Tian, Haoran Ding, Xiao Luo, Joshua Domeyer, Rini Sherony, Zhengming Ding
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 |
|---|---|---|---|
| use InAction (last frame) instead of OIA | OIA · PSI | action overall accuracy: 0.643 -> 0.719 | Improved |
| use InAction (last frame) instead of OIA | OIA · PSI | action mean accuracy: 0.593 -> 0.704 | Improved |
| add temporal input (whole video) to InAction (ablation) | InAction · PSI | action overall accuracy: 0.719 -> 0.734 | Improved |
| add temporal input (whole video) to InAction (ablation) | InAction · PSI | action mean accuracy: 0.704 -> 0.722 | Improved |
| add temporal input (whole video) to InAction (ablation) | InAction · PSI | explanation mean F1: 0.203 -> 0.223 | Improved |
| InAction versus OIA on rare class Stop | OIA · PSI | Stop accuracy: 0.463 -> 0.672 | Improved |
| add local proposal branch to OIA-global (ablation) | OIA-global · PSI | action overall accuracy: 0.635 -> 0.643 | No clear effect |
| InAction versus reproduced OIA* on BDD-OIA | OIA* · BDD-OIA | action overall F1: 0.705 -> 0.714 | Improved |
| InAction versus reported OIA on BDD-OIA action | OIA · BDD-OIA | action overall F1: 0.734 -> 0.714 | Hurt |
| InAction versus reproduced OIA* on BDD-OIA explanation | OIA* · BDD-OIA | explanation overall F1: 0.501 -> 0.565 | Improved |
| InAction versus reproduced OIA* on BDD-OIA explanation | OIA* · BDD-OIA | explanation mean F1: 0.293 -> 0.347 | Improved |
| learn prototypes on global feature map instead of detected proposals (ablation) | InAction · BDD-OIA | action overall F1: 0.706 -> 0.714 | Improved |
Closest papers by meaning
- Towards Safer and Understandable Driver Intention Prediction (ICCV 2025 · similarity 0.90)
- ReAL-AD: Towards Human-Like Reasoning in End-to-End Autonomous Driving (ICCV 2025 · similarity 0.89)
- AutoVLA: A Vision-Language-Action Model for End-to-End Autonomous Driving with Adaptive Reasoning and Reinforcement Fine-Tuning (NeurIPS 2025 · similarity 0.87)
- Continuously Learning, Adapting, and Improving: A Dual-Process Approach to Autonomous Driving (NeurIPS 2024 · similarity 0.87)
- Drive-R1: Bridging Reasoning and Planning in VLMs for Autonomous Driving with Reinforcement Learning (AAAI 2026 · similarity 0.86)
- SGDrive: Scene-to-Goal Hierarchical World Cognition for Autonomous Driving (CVPR 2026 · similarity 0.86)
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.
