REVIEW 7 cited by
Latent Variable Sequential Set Transformers For Joint Multi-Agent Motion Prediction
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Robust multi-agent trajectory prediction is essential for the safe control of robotic systems. A major challenge is to efficiently learn a representation that approximates the true joint distribution of contextual, social, and temporal information to enable planning. We propose Latent Variable Sequential Set Transformers which are encoder-decoder architectures that generate scene-consistent multi-agent trajectories. We refer to these architectures as "AutoBots". The encoder is a stack of interleaved temporal and social multi-head self-attention (MHSA) modules which alternately perform equivariant processing across the temporal and social dimensions. The decoder employs learnable seed parameters in combination with temporal and social MHSA modules allowing it to perform inference over the entire future scene in a single forward pass efficiently. AutoBots can produce either the trajectory of one ego-agent or a distribution over the future trajectories for all agents in the scene. For the single-agent prediction case, our model achieves top results on the global nuScenes vehicle motion prediction leaderboard, and produces strong results on the Argoverse vehicle prediction challenge. In the multi-agent setting, we evaluate on the synthetic partition of TrajNet++ dataset to showcase the model's socially-consistent predictions. We also demonstrate our model on general sequences of sets and provide illustrative experiments modelling the sequential structure of the multiple strokes that make up symbols in the Omniglot data. A distinguishing feature of AutoBots is that all models are trainable on a single desktop GPU (1080 Ti) in under 48h.
Forward citations
Cited by 7 Pith papers
-
Scaling Laws of Motion Forecasting and Planning -- Technical Report
Motion forecasting models improve with compute as a power law, with optimal model size growing 1.5x faster than dataset size, and closed-loop driving failures also decreasing with scale.
-
AutoWorld: Learning Multi-Agent Traffic Simulation with Self-Supervised World Models
AutoWorld learns a self-supervised LiDAR occupancy world model and conditions a diffusion-based motion generator on its forecasts, reporting the top Waymo Sim Agents realism score.
-
Safety Evaluation of Motion Plans Using Trajectory Predictors as Forward Reachable Set Estimators
FORCE-OPT extracts calibrated, multi-modal reachable sets from GMM trajectory predictors using convex optimization and conformal prediction, achieving the lowest balanced error rate in safety evaluation on nuScenes.
-
Goal-based Trajectory Prediction for improved Cross-Dataset Generalization
A graph neural network with staged lane-then-point goal selection cuts the cross-dataset error drop (e.g., b-minFDE6 relative drop 8% vs 20-30%) when moving from Argoverse2 to NuScenes.
-
Generative AI for Testing of Autonomous Driving Systems: A Survey
A systematic survey that organizes 91 studies of generative AI for autonomous driving testing into six scenario-based tasks and catalogs 27 limitations.
-
Surprise Potential as a Measure of Interactivity in Driving Scenarios
A counterfactual surprise metric, Hist-prim with query-centric feedforward prediction and Wasserstein distance, identifies interactive driving scenarios with 0.82+ Spearman correlation to a human-trained reward model.
-
Generative AI for Autonomous Driving: A Review
A review of generative models (VAEs, GANs, diffusion, transformers, LLMs) applied to map generation, scenario generation, trajectory prediction, and motion planning for autonomous driving.
Discussion (0). Continue with ORCID to comment.