Research map / Multi-agent systems and games / Human-AI and goal-conditioned
Latent Variable Sequential Set Transformers for Joint Multi-Agent Motion Prediction
Read the paperCheck 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 joint social attention during future trajectory generation (ablation) | AutoBot · TrajNet++ validation set | Number of collisions: 1144 -> 139 | Helped |
| Add joint social attention during future trajectory generation (ablation) | AutoBot · TrajNet++ validation set | Scene-level Min ADE (6): 0.316 -> 0.128 | Helped |
| Add joint social attention during future trajectory generation (ablation) | AutoBot · TrajNet++ validation set | Scene-level Min FDE (6): 0.431 -> 0.234 | Helped |
| Add joint social attention during future trajectory generation (ablation) | AutoBot · TrajNet++ validation set | Ego Agent's Min ADE (6): 0.098 -> 0.095 | Helped |
| Use an ensemble of three AutoBot-Ego models | AutoBot-Ego · nuScenes | Min ADE (10): 1.05 -> 1.03 | Helped |
| Remove map information (ablation) | AutoBot-Ego · nuScenes | trajectory prediction metrics | Hurt |
| Remove discrete latent variables and decoder seed parameters (ablation) | AutoBot-Ego · nuScenes | ADE | Hurt |
| Replace the seed-parameter decoder with an autoregressive decoder (ablation) | AutoBot · inference speed | inference rate | Hurt |
Closest papers by meaning
- Scene Transformer: A unified architecture for predicting future trajectories of multiple agents (ICLR 2022 · similarity 0.91)
- Multiple Futures Prediction (NeurIPS 2019 · similarity 0.91)
- SmartPretrain: Model-Agnostic and Dataset-Agnostic Representation Learning for Motion Prediction (ICLR 2025 · similarity 0.86)
- SMART: Scalable Multi-agent Real-time Motion Generation via Next-token Prediction (NeurIPS 2024 · similarity 0.86)
- Trajectory Mamba: Efficient Attention-Mamba Forecasting Model Based on Selective SSM (CVPR 2025 · similarity 0.86)
- AutoTraces: Autoregressive Trajectory Forecasting via Multimodal Large Language Models (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.