Search papers, labs, and topics across Lattice.
This paper investigates the alignment of representations in vision-language-action (VLA) policies for contact-rich robotic manipulation, focusing on the role of tactile prediction as a supervisory signal. Through linear probe analysis, the authors reveal that future tactile states are most accurately predicted from intermediate action-expert features rather than from vision-language features or final action states. The proposed Latent Tactile Predictor (LTP) leverages these insights to enhance tactile grounding, resulting in superior performance in real-world manipulation tasks compared to traditional methods.
Future tactile states can be predicted more effectively from intermediate action features, transforming how we approach tactile supervision in robotic manipulation.
Tactile-enhanced vision-language-action (VLA) policies have been introduced for contact-rich manipulation, where critical interaction states are often hidden from vision. Future tactile prediction is a promising way to use touch because it turns tactile outcomes into supervision for action-induced contact dynamics. Yet VLA policies contain representations with different roles, from perceptual encoding to motor prediction, making it unclear where this supervision should be applied. We study this as a representation-alignment problem. Through a linear probe analysis, we find that future tactile states are most predictable from intermediate action-expert features, rather than from vision-language features or final action states. Motivated by this observation, we introduce a lightweight Latent Tactile Predictor (LTP), which predicts compact future tactile embeddings from the identified intermediate representation. By avoiding direct prediction of noisy raw tactile signals, LTP provides an action-outcome grounding signal that aligns intermediate action representations with future contact consequences. Experiments on real-world contact-rich manipulation tasks show that representation-aligned tactile grounding outperforms less aligned or multi-interface tactile prediction, highlighting the importance of where tactile supervision is applied.
