aiqbench analysis
This research introduces a training-free trajectory prediction method that matches the accuracy of large transformer models while eliminating GPU requirements and learned parameters. By leveraging historical state transitions with product kernels, it maintains stability in data-scarce regimes where standard deep learning approaches fail catastrophically. The approach offers a practical pathway for deploying robust multi-modal forecasting in resource-constrained environments without heavy computational overhead.
Non-Parametric Spatiotemporal Trajectory Prediction via State-Conditioned Transition Sampling
arXiv:2608.14349v1 Announce Type: new Abstract: We present a training-free method for multi-modal trajectory prediction that achieves comparable accuracy to a 57M-parameter transformer while requiring no GPU and zero learned parameters. The method builds a transition table of historical state-to-next-position pairs and retrieves neighbors using a product kernel over spatial proximity, bearing, speed, and temporal context. Two inference modes operate over this shared representation: diversity-penalized sampling produces trajectories covering distinct plausible routes, while beam search finds the highest-likelihood path. On the TrAISformer benchmark (Danish Maritime AIS), our method achieves competitive accuracy at full data availability and dramatically outperforms the transformer in data-scarce regimes---remaining stable down to 10% of training data where TrAISformer degrades catastrophically. This enables deployment in new geographic regions from an order of magnitude less historical data, and with no GPU training.
Abstract:We present a training-free method for multi-modal trajectory prediction that achieves comparable accuracy to a 57M-parameter transformer while requiring no GPU and zero learned parameters. The method builds a transition table of historical state-to-next-position pairs and retrieves neighbors using a product kernel over spatial proximity, bearing, speed, and temporal context. Two inference modes operate over this shared representation: diversity-penalized sampling produces trajectories covering distinct plausible routes, while beam search finds the highest-likelihood path. On the TrAISformer benchmark (Danish Maritime AIS), our method achieves competitive accuracy at full data availability and dramatically outperforms the transformer in data-scarce regimes---remaining stable down to 10% of training data where TrAISformer degrades catastrophically. This enables deployment in new geographic regions from an order of magnitude less historical data, and with no GPU training.
Machine Learning (cs.LG)
arXiv:2608.14349 [cs.LG]
(or arXiv:2608.14349v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2608.14349 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 222 words
Related coverage
An Empirical Markov Chain Car-Following (MC-CF) Model
arXiv · Sep 14, 2026
SPIBER: Reconstructing Free Energy Landscapes from Short, Unconverged Trajectories with Generative Flow Networks
arXiv · Sep 22, 2026
Untrusted advice for AI control: Short, strong advice significantly uplifts weak LLMs
LessWrong · Jul 27, 2026
The Entangled Dimensions of Decision Theory
LessWrong · Jul 30, 2026