Latent Action Learning Requires Supervision in the Presence of Distractors
This addresses a critical bottleneck for embodied AI by enabling more robust latent action learning from noisy video data, though it is incremental as it modifies an existing method.
The paper tackles the problem of latent action learning failing in real-world videos with distractors, showing that a modified method (LAOM) improves latent action quality by 8x and that adding minimal supervision (2.5% of data) boosts downstream performance by 4.2x.
Recently, latent action learning, pioneered by Latent Action Policies (LAPO), have shown remarkable pre-training efficiency on observation-only data, offering potential for leveraging vast amounts of video available on the web for embodied AI. However, prior work has focused on distractor-free data, where changes between observations are primarily explained by ground-truth actions. Unfortunately, real-world videos contain action-correlated distractors that may hinder latent action learning. Using Distracting Control Suite (DCS) we empirically investigate the effect of distractors on latent action learning and demonstrate that LAPO struggle in such scenario. We propose LAOM, a simple LAPO modification that improves the quality of latent actions by 8x, as measured by linear probing. Importantly, we show that providing supervision with ground-truth actions, as few as 2.5% of the full dataset, during latent action learning improves downstream performance by 4.2x on average. Our findings suggest that integrating supervision during Latent Action Models (LAM) training is critical in the presence of distractors, challenging the conventional pipeline of first learning LAM and only then decoding from latent to ground-truth actions.