paper-with-me

홈 › Papers

Run-time Observation Interventions Make Vision-Language-Action Models More Visually Robust

2024-10-02 · Asher J. Hancock, Allen Z. Ren, Anirudha Majumdar

Vision-language-action (VLA) models trained on large-scale internet data and robot demonstrations have the potential to serve as generalist robot policies. However, despite their large-scale training, VLAs are often brittle to task-irrelevant visual details such as distractor objects or background colors. We introduce Bring Your Own VLA (BYOVLA): a run-time intervention scheme that (1) dynamically identifies regions of the input image that the model is sensitive to, and (2) minimally alters task-irrelevant regions to reduce the model's sensitivity using automated image editing tools. Our approach is compatible with any off the shelf VLA without model fine-tuning or access to the model's weights. Hardware experiments on language-instructed manipulation tasks demonstrate that BYOVLA enables state-of-the-art VLA models to nearly retain their nominal performance in the presence of distractor objects and backgrounds, which otherwise degrade task success rates by up to 40%. Website with additional information, videos, and code: https://aasherh.github.io/byovla/ .

📄 PDF Abstract BibTeX arXiv:2410.01971

Code (0)

등록된 구현이 없습니다.

Tasks

Vision-Language-Action

Similar Papers 제목 키워드 기반

Bayesian causal inference via probabilistic program synthesis

2019-10-30 · Sam Witty, Alexander Lew, David Jensen, Vikash Mansinghka

Causal inference can be formalized as Bayesian inference that combines a prior distribution over causal models and likelihoods that account for both observations and interventions. We show that it is possible to implemen…

Bayesian InferenceCausal InferenceProbabilistic ProgrammingProgram Synthesis

Beyond Predictions in Neural ODEs: Identification and Interventions

2021-06-23 · Hananeh Aliee, Fabian J. Theis, Niki Kilbertus

Spurred by tremendous success in pattern matching and prediction tasks, researchers increasingly resort to machine learning to aid original scientific discovery. Given large amounts of observational data about a system, …

scientific discoveryTime SeriesTime Series Analysis

Learning and Sampling of Atomic Interventions from Observations

2020-02-11 · ICML 2020 1 · Arnab Bhattacharyya, Sutanu Gayen, Saravanan Kandasamy, Ashwin Maran 외

We study the problem of efficiently estimating the effect of an intervention on a single variable (atomic interventions) using observational samples in a causal Bayesian network. Our goal is to give algorithms that are e…

CG-World: A Large-Scale World-State Dataset and Protocol for World Models

2026-07-29 · Yiming Cai, Fangjie Yu, Meiqing Yu, Ziyue Shi 외 arxiv

World models must learn the joint dynamics of states, actions, events, and observations, yet existing video, robotics, and simulation datasets usually capture only part of this structure. We introduce CG-World, a large-s…

Video Generation

SceneForge: Structured World Supervision from 3D Interventions

2026-05-14 · Jizhizi Li, Jiayang Ao, Danny Wicks, Petru-Daniel Tudosiu arxiv

Many multimodal learning tasks require supervision that remains consistent across edits, viewpoints, and scene-level interventions. However, such supervision is difficult to obtain from observation-level datasets, which …