AtomBridge: Agentic VLA Inference Plugin for Long-Horizon Tasks in Scientific Experiments

작성자

카테고리:

← 피드로
arXiv cs.AI · Yiwen Pang, Bo Zhou, Changjin Li, Xuanhao Wang, Shengxiang Xu, Deng-Bao Wang, Peng Cheng, Shimin Di, Jingkuan Song, Min-Ling Zhang · 2026-08-17 AI

[Submitted on 10 Feb 2026 (v1), last revised 14 Aug 2026 (this version, v2)]

View PDF HTML (experimental)

Abstract:Robotic laboratories play a critical role in autonomous scientific discovery by enabling scalable, continuous experimental execution. Recent vision-language-action (VLA) models offer a promising foundation for robotic laboratories. However, scientific experiments typically involve long-horizon tasks composed of multiple atomic tasks. Existing VLA models may fail to perform composed tasks formed by reordering and composing these known atomic actions. This limitation can arise from a skill-chaining gap caused by robot-state mismatch: the terminal robot state of one skill can fall outside the valid initial-state distribution of the next. To address this challenge, we propose AtomBridge, an Agentic VLA Inference Plugin for Long-Horizon Tasks in Scientific Experiments. AtomBridge attaches at inference time to a VLA policy already fine-tuned on atomic tasks, while keeping its weights fixed. At each task boundary, it uses LLM-based transition reasoning and robotic-action code generation to insert transitional actions between consecutive tasks. This plug-and-play design mitigates the skill-chaining gap caused by robot-state mismatch without additional VLA fine-tuning or demonstrations of composed long-horizon sequences. Across scientific manipulation sequences in simulation and a real-world experimental environment, AtomBridge improves execution continuity and per-step atomic-task success. On 8-step composed tasks, AtomBridge improves full-sequence success by 10%~25%.

Submission history

From: Yiwen Pang [view email]
[v1] Tue, 10 Feb 2026 05:50:19 UTC (11,328 KB)
[v2] Fri, 14 Aug 2026 06:48:51 UTC (3,287 KB)

원문에서 계속 ↗

추출 본문 · 출처: arxiv.org · https://arxiv.org/abs/2602.09430