Much Ado About Noising: Dispelling the Myths of Generative Robotic Control
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866910029686243328 |
|---|---|
| author | Pan, Chaoyi Anantharaman, Giri Huang, Nai-Chieh Jin, Claire Pfrommer, Daniel Yuan, Chenyang Permenter, Frank Qu, Guannan Boffi, Nicholas Shi, Guanya Simchowitz, Max |
| author_facet | Pan, Chaoyi Anantharaman, Giri Huang, Nai-Chieh Jin, Claire Pfrommer, Daniel Yuan, Chenyang Permenter, Frank Qu, Guannan Boffi, Nicholas Shi, Guanya Simchowitz, Max |
| contents | Generative models, like flows and diffusions, have recently emerged as popular and efficacious policy parameterizations in robotics. There has been much speculation as to the factors underlying their successes, ranging from capturing multi-modal action distribution to expressing more complex behaviors. In this work, we perform a comprehensive evaluation of popular generative control policies (GCPs) on common behavior cloning (BC) benchmarks. We find that GCPs do not owe their success to their ability to capture multi-modality or to express more complex observation-to-action mappings. Instead, we find that their advantage stems from iterative computation, as long as intermediate steps are supervised during training and this supervision is paired with a suitable level of stochasticity. As a validation of our findings, we show that a minimum iterative policy (MIP), a lightweight two-step regression-based policy, essentially matches the performance of flow GCPs, and often outperforms distilled shortcut models. Our results suggest that the distribution-fitting component of GCPs is less salient than commonly believed, and point toward new design spaces focusing solely on control performance. Project page: https://simchowitzlabpublic.github.io/much-ado-about-noising-project/ |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2512_01809 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Much Ado About Noising: Dispelling the Myths of Generative Robotic Control Pan, Chaoyi Anantharaman, Giri Huang, Nai-Chieh Jin, Claire Pfrommer, Daniel Yuan, Chenyang Permenter, Frank Qu, Guannan Boffi, Nicholas Shi, Guanya Simchowitz, Max Robotics Machine Learning Generative models, like flows and diffusions, have recently emerged as popular and efficacious policy parameterizations in robotics. There has been much speculation as to the factors underlying their successes, ranging from capturing multi-modal action distribution to expressing more complex behaviors. In this work, we perform a comprehensive evaluation of popular generative control policies (GCPs) on common behavior cloning (BC) benchmarks. We find that GCPs do not owe their success to their ability to capture multi-modality or to express more complex observation-to-action mappings. Instead, we find that their advantage stems from iterative computation, as long as intermediate steps are supervised during training and this supervision is paired with a suitable level of stochasticity. As a validation of our findings, we show that a minimum iterative policy (MIP), a lightweight two-step regression-based policy, essentially matches the performance of flow GCPs, and often outperforms distilled shortcut models. Our results suggest that the distribution-fitting component of GCPs is less salient than commonly believed, and point toward new design spaces focusing solely on control performance. Project page: https://simchowitzlabpublic.github.io/much-ado-about-noising-project/ |
| title | Much Ado About Noising: Dispelling the Myths of Generative Robotic Control |
| topic | Robotics Machine Learning |
| url | https://arxiv.org/abs/2512.01809 |