Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Foresight in Motion: Reinforcing Trajectory Prediction with Reward Heuristics
Muleilan Pei, Shaoshuai Shi, Xuesong Chen, Xu Liu, Shaojie Shen
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 reward-driven intention reasoning (QIRL) vs vanilla without it (ablation) | FiM · Argoverse 1 val | brier-minFDE6: 2.879 -> 1.602 | Improved |
| use QIRL instead of cross-attention block (C.A.) for intention priors (ablation) | FiM · Argoverse 1 val | brier-minFDE6: 2.132 -> 1.602 | Improved |
| add reasoning, minFDE6 (ablation) | FiM · Argoverse 1 val | minFDE6: 2.185 -> 1.008 | Improved |
| add refinement to OGM-only model (ablation) | FiM · Argoverse 1 val | brier-minFDE6: 1.670 -> 1.602 | Improved |
| add auxiliary OGM head to refinement-only model (ablation) | FiM · Argoverse 1 val | brier-minFDE6: 1.801 -> 1.602 | Improved |
| add Bi-Mamba to MLP decoder (ablation) | FiM decoder · Argoverse 1 val | brier-minFDE6: 1.720 -> 1.649 | Improved |
| add mode self-attention to MLP decoder (ablation) | FiM decoder · Argoverse 1 val | brier-minFDE6: 1.720 -> 1.682 | Improved |
| combine Bi-Mamba with self-attention (ablation) | FiM decoder · Argoverse 1 val | Brier score: 0.622 -> 0.594 | Improved |
| use bidirectional Mamba with two CLS tokens instead of unidirectional (ablation) | FiM decoder · Argoverse 1 val | brier-minFDE6: 1.636 -> 1.602 | Improved |
| increase Mamba layers from 4 to 8 (ablation) | FiM decoder · Argoverse 1 val | brier-minFDE6: 1.602 -> 1.628 | Hurt |
| use only 2 Mamba layers instead of 4 (ablation) | FiM decoder · Argoverse 1 val | brier-minFDE6: 1.602 -> 1.635 | Hurt |
| longer future supervision for GRT (30 to 60 timestamps) (ablation) | FiM · Argoverse 2 customized val | Brier score: 0.617 -> 0.603 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- DeMo — DeMo: Decoupling Motion Forecasting into Directional Intentions and Dynamic States (NeurIPS 2024) · better on 2 of 2 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- DeMo: Decoupling Motion Forecasting into Directional Intentions and Dynamic States (NeurIPS 2024 · similarity 0.90)
- GoIRL: Graph-Oriented Inverse Reinforcement Learning for Multimodal Trajectory Prediction (ICML 2025 · similarity 0.88)
- DriveGPT: Scaling Autoregressive Behavior Models for Driving (ICML 2025 · similarity 0.88)
- Drive-R1: Bridging Reasoning and Planning in VLMs for Autonomous Driving with Reinforcement Learning (AAAI 2026 · similarity 0.88)
- Future-Aware End-to-End Driving: Bidirectional Modeling of Trajectory Planning and Scene Evolution (NeurIPS 2025 · similarity 0.87)
- UniMotion: A Unified Motion Framework for Simulation, Prediction and Planning (NeurIPS 2025 · similarity 0.87)
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.
