Research map / Multi-agent systems and games / Human-AI and goal-conditioned
DONUT: A Decoder-Only Model for Trajectory Prediction
Markus Knoche, Daan de Geus, Bastian Leibe
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 decoder-only instead of encoder-decoder (ablation) | QCNet baseline · Argoverse 2 val | b-minFDE6: 1.874 -> 1.838 | Improved |
| use decoder-only instead of encoder-decoder (ablation) | QCNet baseline · Argoverse 2 val | minFDE6: 1.253 -> 1.198 | Improved |
| use decoder-only instead of encoder-decoder (ablation) | QCNet baseline · Argoverse 2 val | MR6: 0.157 -> 0.145 | Improved |
| use decoder-only instead of encoder-decoder (ablation) | QCNet baseline · Argoverse 2 val | minADE6: 0.720 -> 0.745 | Hurt |
| add overprediction to decoder-only (ablation) | DONUT w/o refinement · Argoverse 2 val | b-minFDE6: 1.838 -> 1.838 | No clear effect |
| add refinement to decoder-only (ablation) | DONUT w/o overprediction · Argoverse 2 val | b-minFDE6: 1.838 -> 1.835 | No clear effect |
| add refinement to decoder-only (ablation) | DONUT w/o overprediction · Argoverse 2 val | minFDE6: 1.198 -> 1.218 | Hurt |
| add overprediction on top of decoder-only + refinement (ablation) | DONUT · Argoverse 2 val | b-minFDE6: 1.835 -> 1.807 | Improved |
| add refinement on top of decoder-only + overprediction (ablation) | DONUT · Argoverse 2 val | b-minFDE6: 1.838 -> 1.807 | Improved |
| add overprediction on top of decoder-only + refinement (ablation) | DONUT · Argoverse 2 val | minFDE6: 1.218 -> 1.176 | Improved |
| add overprediction on top of decoder-only + refinement (ablation) | DONUT · Argoverse 2 val | MR6: 0.150 -> 0.144 | Improved |
Later papers that compared with this method
1 later accepted paper reports DONUT as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.
- ECTraj: Enhanced Consistency Training for Multi-Agent Trajectory Prediction (ECCV 2026 · better on 1 of 1 shared result)
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 0 of 1 on most shared results.
- DeMo — DeMo: Decoupling Motion Forecasting into Directional Intentions and Dynamic States (NeurIPS 2024) · better on 3 of 7 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
- AutoTraces: Autoregressive Trajectory Forecasting via Multimodal Large Language Models (CVPR 2026 · similarity 0.89)
- Trajectory Mamba: Efficient Attention-Mamba Forecasting Model Based on Selective SSM (CVPR 2025 · similarity 0.89)
- DeMo: Decoupling Motion Forecasting into Directional Intentions and Dynamic States (NeurIPS 2024 · similarity 0.88)
- Trajeglish: Traffic Modeling as Next-Token Prediction (ICLR 2024 · similarity 0.88)
- Motion Forecasting in Continuous Driving (NeurIPS 2024 · similarity 0.87)
- SmartPretrain: Model-Agnostic and Dataset-Agnostic Representation Learning for Motion Prediction (ICLR 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.
