# Generative Action-Chunk Sampling for Adaptive Stiffness Control in Physical Human-Robot Collaboration

arXiv: https://arxiv.org/abs/2608.25284 · HTML: https://mlsanity.com/paper/2608.25284/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 17 of its 30 references that are in the map.
- First posted on arXiv 26 Aug 2026.

## Builds on (most cited first, 17 of 17)

- [An Image is Worth 16x16 Words: Transformers for Image Recognition at Scale](https://mlsanity.com/paper/2010.11929/) (2020)
- [Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardware](https://mlsanity.com/paper/2304.13705/) (2023)
- [End-to-End Training of Deep Visuomotor Policies](https://mlsanity.com/paper/1504.00702/) (2015)
- [Semi-Supervised Learning with Deep Generative Models](https://mlsanity.com/paper/1406.5298/) (2014)
- [Causal Confusion in Imitation Learning](https://mlsanity.com/paper/1905.11979/) (2019)
- [Real-Time Execution of Action Chunking Flow Policies](https://mlsanity.com/paper/2506.07339/) (2025)
- [Making Sense of Vision and Touch: Self-Supervised Learning of Multimodal Representations for Contact-Rich Tasks](https://mlsanity.com/paper/1810.10191/) (2018)
- [A Review of Robot Learning for Manipulation: Challenges, Representations, and Algorithms](https://mlsanity.com/paper/1907.03146/) (2019)
- [An Unbiased Look at Datasets for Visuo-Motor Pre-Training](https://mlsanity.com/paper/2310.09289/) (2023)
- [FACTR: Force-Attending Curriculum Training for Contact-Rich Policy Learning](https://mlsanity.com/paper/2502.17432/) (2025)
- [Variable impedance control and learning -- A review](https://mlsanity.com/paper/2010.06246/) (2020)
- [A Review of Prospects and Opportunities in Disassembly with Human-Robot Collaboration](https://mlsanity.com/paper/2310.13643/) (2023)
- [FTACT: Force Torque aware Action Chunking Transformer for Pick-and-Reorient Bottle Task](https://mlsanity.com/paper/2509.23112/) (2025)
- [Occlusion-Aware 3D Hand-Object Pose Estimation with Masked AutoEncoders](https://mlsanity.com/paper/2506.10816/) (2025)
- [DTRT: Enhancing Human Intent Estimation and Role Allocation for Physical Human-Robot Collaboration](https://mlsanity.com/paper/2505.17490/) (2025)
- [Haptic Communication in Human-Human and Human-Robot Co-Manipulation](https://mlsanity.com/paper/2509.18327/) (2025)
- [OmniVIC: A Self-Improving Variable Impedance Controller with Vision-Language In-Context Learning for Safe Robotic Manipulation](https://mlsanity.com/paper/2510.17150/) (2025)

## Abstract (arXiv)

Physical human-robot collaboration requires a robot to provide assistance when human intention is clear while remaining compliant when several future motions are plausible. We present an adaptive stiffness framework based on generative action-chunk sampling. Conditioned on an RGB image and external joint-torque estimates, the policy samples multiple latent variables from an observation-conditioned prior and decodes them into future action chunks. Variation among the sampled action chunks is used to continuously adapt joint stiffness and damping. Greater variation makes the robot more compliant to facilitate human guidance, whereas lower variation provides firmer assistance. In a real-world collaborative transport task with four possible directions, the proposed method achieved an average success rate of 0.95, compared with 0.83 for a fixed-stiffness ablation and 0.69 for a deterministic baseline. Near direction determination, variation among the sampled action chunks increased, and the controller reduced stiffness accordingly. These results suggest that variation among actions sampled by a generative policy can serve as an online control signal for balancing assistance and compliance in physical human-robot interaction.