Revisiting Numerical Forecasting Models for Language-Based Trajectory Prediction

Key points

• This formulation enables the model to capture behavioral intent and social context beyond coordinate dynamics alone.1
• To address this limitation, we introduce MoRE (Mixture of Reward Experts), a refinement framework that transfers numerical forecasting priors into a pretrained language-based predictor through reinforcement learning.1
• Five frozen numerical predictors provide complementary coordinate-level knowledge of motion and interactions.1
• A ground-truth reward anchors the prediction to the target trajectory.1

Sources (1)

• [1]Revisiting Numerical Forecasting Models for Language-Based Trajectory Prediction
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 08:28 AM

  Language-based trajectory predictors represent coordinates as discrete tokens and learn auxiliary tasks such as destination and group reasoning.

  This formulation enables the model to capture behavioral intent and social context beyond coordinate dynamics alone.

Extractive summary: sentences quoted from the sources.