LADA: Low-Cost Driving Annotations via Latent Action Models

abursuc · x · 2026-09-17

Presented at the #ssad2026 workshop, LADA (Latent Action Driving Annotations) is a low-cost strategy for generating language annotations for driving data. Built on latent action models (LAMs), it learns to produce latent actions from only a few observation-language pairs, imitating the function of language to annotate driving datasets cheaply. The post also includes a primer on LAMs.

Related event: LADA: Low-Cost Language Annotations for Autonomous Driving via Latent Actions(2 posts)→

Original post →

More from Research

Research channel →