Pith. sign in

REVIEW 9 cited by

Tactile Beyond Pixels: Multisensory Touch Representations for Robot Manipulation

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2506.14754 v1 pith:VCABCPFV submitted 2025-06-17 cs.RO

Tactile Beyond Pixels: Multisensory Touch Representations for Robot Manipulation

classification cs.RO
keywords sparsh-xtouchtactilemanipulationmultisensoryphysicalpropertiesrepresentations
verification ladder T0 review T1 audit T2 compute T3 formal T4 reserved
0 comments
read the original abstract

We present Sparsh-X, the first multisensory touch representations across four tactile modalities: image, audio, motion, and pressure. Trained on ~1M contact-rich interactions collected with the Digit 360 sensor, Sparsh-X captures complementary touch signals at diverse temporal and spatial scales. By leveraging self-supervised learning, Sparsh-X fuses these modalities into a unified representation that captures physical properties useful for robot manipulation tasks. We study how to effectively integrate real-world touch representations for both imitation learning and tactile adaptation of sim-trained policies, showing that Sparsh-X boosts policy success rates by 63% over an end-to-end model using tactile images and improves robustness by 90% in recovering object states from touch. Finally, we benchmark Sparsh-X ability to make inferences about physical properties, such as object-action identification, material-quantity estimation, and force estimation. Sparsh-X improves accuracy in characterizing physical properties by 48% compared to end-to-end approaches, demonstrating the advantages of multisensory pretraining for capturing features essential for dexterous manipulation.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 9 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Tactile Genesis: Exploring Tactile Sensors at Scale for Learning Dexterous Tasks

    cs.RO 2026-06 unverdicted novelty 7.0

    Tactile Genesis provides a scalable multi-type tactile simulator and ablation results showing whole-hand coverage with per-taxel force/torque sensing outperforms fingertip-only or other modalities across three dextero...

  2. Tactile Genesis: Exploring Tactile Sensors at Scale for Learning Dexterous Tasks

    cs.RO 2026-06 conditional novelty 6.5

    Whole-hand tactile coverage and per-taxel force/torque dominate sensor type and resolution for learning three dexterous tasks in a new high-throughput tactile simulator.

  3. Representation-Aligned Tactile Grounding for Contact-Rich Robotic Manipulation

    cs.RO 2026-07 conditional novelty 6.0

    Future tactile prediction applied to intermediate action-expert features, rather than visual-language or final-action features, improves contact-rich manipulation in SmolVLA and π0.

  4. TactX: Learning Shared Tactile Representations Across Diverse Sensors

    cs.RO 2026-06 unverdicted novelty 6.0

    TactX learns a shared latent representation across three tactile sensor modalities via joint training on paired contacts, enabling zero-shot policy transfer and higher success on pick-and-place, insertion, wiping, and...

  5. Heterogeneous Tactile Transformer

    cs.RO 2026-06 unverdicted novelty 6.0

    HTT learns shared representations across heterogeneous tactile sensors using a new paired dataset and pretraining objectives, enabling transfer to unseen sensors and tasks.

  6. Multi-Resolution Tactile Imitation Learning for Contact-Rich Robotic Manipulation

    cs.RO 2026-06 unverdicted novelty 6.0

    MiTaS fuses multi-resolution tactile data from GelSight and Evetac sensors with vision using modality-specific stems and transformer fusion to condition flow-matching policies, reporting 80% average success on five co...

  7. TacO: Benchmarking Tactile Sensors for Object Manipulation

    cs.RO 2026-05 unverdicted novelty 6.0

    The paper provides a task-driven benchmark comparing visual, acoustic, magnetic, and resistive tactile sensors on three manipulation tasks and concludes that sensor utility depends on modality, material friction, and ...

  8. Online World Modeling Enables Real-World Inverse Reinforcement Learning from Observation

    cs.RO 2026-02 conditional novelty 6.0

    MPAIL2 demonstrates real-world manipulation learning from observation alone, without rewards or action labels, plus positive online transfer.

  9. DECO: Decoupled Multimodal Diffusion Transformer for Bimanual Dexterous Manipulation with a Plugin Tactile Adapter

    cs.RO 2026-02 conditional novelty 6.0

    A decoupled multimodal diffusion transformer with a LoRA tactile adapter improves real-world bimanual manipulation success by 21 percentage points over a diffusion-policy baseline; a new 50-hour tactile bimanual datas...