# Tactile Curiosity Drives Robot Interaction

arXiv: https://arxiv.org/abs/2609.40134 · HTML: https://mlsanity.com/paper/2609.40134/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in RL for robot locomotion and manipulation, one of 15 directions in Deep reinforcement learning methods.
- Builds on 33 of its 41 references that are in the map.
- First posted on arXiv 30 Sep 2026.

## Placed in

- Deep reinforcement learning methods / [RL for robot locomotion and manipulation](https://mlsanity.com/deeprl/rl-for-robot-locomotion-and-manipulation/)

## Builds on (most cited first, 30 of 33)

- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Playing Atari with Deep Reinforcement Learning](https://mlsanity.com/paper/1312.5602/) (2013)
- [Addressing Function Approximation Error in Actor-Critic Methods](https://mlsanity.com/paper/1802.09477/) (2018)
- [Deep Reinforcement Learning with Double Q-learning](https://mlsanity.com/paper/1509.06461/) (2015)
- [LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learning](https://mlsanity.com/paper/2306.03310/) (2023)
- [Soft Actor-Critic Algorithms and Applications](https://mlsanity.com/paper/1812.05905/) (2018)
- [Curiosity-driven Exploration by Self-supervised Prediction](https://mlsanity.com/paper/1705.05363/) (2017)
- [Exploration by Random Network Distillation](https://mlsanity.com/paper/1810.12894/) (2018)
- [DeepMind Control Suite](https://mlsanity.com/paper/1801.00690/) (2018)
- [Image Augmentation Is All You Need: Regularizing Deep Reinforcement Learning from Pixels](https://mlsanity.com/paper/2004.13649/) (2020)
- [Planning to Explore via Self-Supervised World Models](https://mlsanity.com/paper/2005.05960/) (2020)
- [Self-Supervised Exploration via Disagreement](https://mlsanity.com/paper/1906.04161/) (2019)
- [Steering Your Diffusion Policy with Latent Space Reinforcement Learning](https://mlsanity.com/paper/2506.15799/) (2025)
- [More Than a Feeling: Learning to Grasp and Regrasp using Vision and Touch](https://mlsanity.com/paper/1805.11085/) (2018)
- [Rotating without Seeing: Towards In-hand Dexterity through Touch](https://mlsanity.com/paper/2303.10880/) (2023)
- [Tactile-RL for Insertion: Generalization to Objects of Unknown Geometry](https://mlsanity.com/paper/2104.01167/) (2021)
- [Beyond Sight: Finetuning Generalist Robot Policies with Heterogeneous Sensors via Language Grounding](https://mlsanity.com/paper/2501.04693/) (2025)
- [Manipulation by Feel: Touch-Based Control with Deep Predictive Models](https://mlsanity.com/paper/1903.04128/) (2019)
- [Optimistic Active Exploration of Dynamical Systems](https://mlsanity.com/paper/2306.12371/) (2023)
- [MaxInfoRL: Boosting exploration in reinforcement learning through information gain maximization](https://mlsanity.com/paper/2412.12098/) (2024)
- [An information-theoretic perspective on intrinsic motivation in reinforcement learning: a survey](https://mlsanity.com/paper/2209.08890/) (2022)
- [TacSL: A Library for Visuotactile Sensor Simulation and Learning](https://mlsanity.com/paper/2408.06506/) (2024)
- [Learning Gentle Object Manipulation with Curiosity-Driven Deep Reinforcement Learning](https://mlsanity.com/paper/1903.08542/) (2019)
- [DexTouch: Learning to Seek and Manipulate Objects with Tactile Dexterity](https://mlsanity.com/paper/2401.12496/) (2024)
- [Improved Learning of Robot Manipulation Tasks via Tactile Intrinsic Motivation](https://mlsanity.com/paper/2102.11051/) (2021)
- [Ground truth force distribution for learning-based tactile sensing: a finite element approach](https://mlsanity.com/paper/1909.04000/) (2019)
- [SOMBRL: Scalable and Optimistic Model-Based RL](https://mlsanity.com/paper/2511.20066/) (2025)
- [Sim-to-real for high-resolution optical tactile sensing: From images to 3D contact force distributions](https://mlsanity.com/paper/2012.11295/) (2020)
- [Touch-based Curiosity for Sparse-Reward Tasks](https://mlsanity.com/paper/2104.00442/) (2021)
- [Tactile-based Object Retrieval From Granular Media](https://mlsanity.com/paper/2402.04536/) (2024)

## Abstract (arXiv)

Mastering robot manipulation skills via reinforcement learning (RL) remains largely sample-inefficient. The most common RL algorithms rely on random action sampling to discover new strategies, resulting in agents that allocate most of their training budget to motions in free space, away from the contacts from which manipulation skills emerge. Existing intrinsic motivation methods based on model disagreement or epistemic uncertainty improve on isotropic noise, but they can also reward uncertainty in functionally irrelevant transitions, such as erratic motions in free space. In this work, we argue that tactile feedback provides a natural signal for exploration, and introduce TacEx, a framework that incorporates touch into epistemic uncertainty-driven exploration by decomposing model uncertainty across sensory modalities and directing curiosity toward the tactile channel. By anchoring curiosity to the sense of touch, TacEx drives the robot to discover complex contact dynamics, learning to manipulate and grasp objects without task rewards or expert demonstrations during exploration. The interaction-dense dataset collected through this tactile-driven curiosity supports offline learning of downstream pick-and-place policies without additional environment interaction. We further use tactile-driven exploration to post-train vision-language-action (VLA) models. Although the VLAs are initially pre-trained without tactile feedback, post-training with TacEx substantially improves downstream performance while remaining highly sample-efficient.