AI Paper Atlas

Research map / Multi-agent systems and games / Human-AI and goal-conditioned

SWIFT: A Small-World Interaction Framework for Flow-Aware Trajectory Prediction in Autonomous Driving

Chengyue Wang, Bin Rao, Haicheng Liao, Bo Wang, Chengzhong Xu, Zhenning Li

TPAMI 2026 · Multi-agent systems and games › Human-AI and goal-conditioned · approach: Data-driven traffic simulation · 0 citations (OpenAlex)

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 changedWhereMeasuredEffect
add Interaction Reasoning module (ablation)SWIFT · nuScenesminFDE_1: 7.37 -> 6.73Improved
add Dynamic Encoder (ablation)SWIFT · nuScenesminADE_1: 3.26 -> 2.89Improved
add Static Encoder (ablation)SWIFT · nuScenesminADE_5: 1.49 -> 1.15Improved
add attention fusion instead of concatenation in hybrid decoder (ablation)SWIFT · nuScenesminFDE_1: 7.49 -> 6.73Improved
add attention fusion instead of addition in hybrid decoder (ablation)SWIFT · nuScenesminADE_5: 1.29 -> 1.15Improved
add Small-world Interaction Network to baseline trained on 60% data (ablation)SWIFT · nuScenes (60% train)minADE_5: 1.35 -> 1.27Improved
add Flow Regime Encoder on top of SWIN, 60% data (ablation)SWIFT · nuScenes (60% train)minADE_5: 1.27 -> 1.23Improved
add FRE and SWIN under noise rho=6 (ablation)SWIFT · nuScenes (noise rho=6)minFDE_1: 8.36 -> 7.26Improved
increase connection-cost exponent from 1 to 3 (ablation)SWIFT · NGSIMRMSE at 5 s: 2.84 -> 2.75Improved
phi=1 versus phi=3 at 1 s horizon (ablation)SWIFT · NGSIMRMSE at 1 s: 0.31 vs 0.34Mixed
lower trajectory loss weight to 0.5 with lambda_b and lambda_c at 0.5 (ablation)SWIFT · MoCADRMSE at 5 s: 2.36 -> 2.62Hurt
lower regime and interaction loss weights from 0.5 to 0.1 (ablation)SWIFT · MoCADRMSE at 5 s: 2.36 -> 2.47Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy