arXiv:2606. 31943v1 Announce Type: new Abstract: Is the sense of touch a mechanism for human babies' learning of visual concepts?
arXiv:2504. 17939v2 Announce Type: replace-cross Abstract: We present a computational model of the mechanisms that may determine infant behavior in the "mobile paradigm".
arXiv:2603. 14604v2 Announce Type: replace-cross Abstract: We propose TacFiLM, a lightweight modality-fusion approach that integrates visual-tactile signals into vision-language-action (VLA) models.
arXiv:2609.23352v1 Announce Type: new Abstract: Bimanual interaction produces complementary tactile views of the same physical process, yet existing tactile representation learning largely models the...
arXiv:2608.28664v1 Announce Type: cross Abstract: Despite the success of foundation models in language and vision, their expansion into embodied AI is bottlenecked by a lack of generalized touch sens...
arXiv:2609.24976v1 Announce Type: cross Abstract: Dexterous manipulation depends on contact dynamics that are often only partially observable from vision. Recent World-Action Models (WAMs) couple pre...
TouchSight is a monocular egocentric vision framework that predicts dense full-hand contact forces from video. It uses 500 hours of pressure‑glove recordings and hand‑object interaction data, and introduces TwinTouch‑20H, a dataset of 20 hours of paired visual data where generative models render gloved recordings as bare‑hand observations while preserving tactile labels. The system outperforms prior methods on OakInk2, generalizes qualitatively to natural bare‑hand egocentric videos from unseen datasets, and improves consistently as glove supervision scales.
arXiv:2609.14783v1 Announce Type: cross Abstract: Robots need touch to manipulate objects safely and reliably, as many properties, such as softness, texture, and contact stability, are hard to infer...
TouchSight is a monocular egocentric vision framework that predicts dense full-hand contact forces without tactile sensors. It uses 500 hours of pressure-glove data and a 20-hour TwinTouch-20H dataset where generative models render gloved recordings as bare-hand videos, bridging the appearance gap. The system outperforms previous methods on OakInk2, generalizes to unseen natural bare-hand egocentric videos, and improves as glove supervision increases.
arXiv:2606. 31694v1 Announce Type: cross Abstract: For robots manipulating open-world objects, tactile representations must generalize to unseen materials.
arXiv:2606. 11637v1 Announce Type: new Abstract: Touch is a key modality for embodied agents to understand the physical world.
VT-MUSE is a multimodal unified sequential representation learning framework for visuotactile manipulation. It uses a two‑stage approach: first, modality‑specific encoders are jointly adapted with cross‑modal temporal alignment and masked‑view consistency; second, a conditional variational latent model processes masked visual sequences and full tactile histories, with auxiliary decoders reconstructing recent visual observations and predicting tactile depth changes. The resulting representation is fed into a lightweight Transformer policy via gated cross‑attention, achieving an 11‑percentage‑point improvement over the strongest baseline in simulation and significant gains in real‑world experiments.
arXiv:2606. 09243v1 Announce Type: cross Abstract: Estimating full-hand grasp pressure from egocentric video is critical for immersive VR and robotic manipulation, yet dense tactile sensing often relies on intrusive hardware.