General-purpose robotic systems operating in open-world environments must achieve both broad generalization and high-precision action execution, a combination that remains challenging for existing Vision-Language-Action (VLA) models. While large Vision-Language Models (VLMs) improve semantic generalization, insufficient embodied reasoning leads to brittle behavior, and conversely, strong reasoning alone is inadequate without precise control. To provide a decoupled and quantitative assessment of this bottleneck, we introduce Embodied Reasoning Intelligence Quotient (ERIQ), a large-scale embodied reasoning benchmark in robotic manipulation, comprising 6K+ question-answer pairs across four reasoning dimensions. By decoupling reasoning from execution, ERIQ enables systematic evaluation and reveals a strong positive correlation between embodied reasoning capability and end-to-end VLA generalization. To bridge the gap from reasoning to precise execution, we propose FACT, a flow-matching-based action tokenizer that converts continuous control into discrete sequences while preserving high-fidelity trajectory reconstruction. The resulting GenieReasoner jointly optimizes reasoning and action in a unified space, outperforming both continuous-action and prior discrete-action baselines in real-world tasks. Together, ERIQ and FACT provide a principled framework for diagnosing and overcoming the reasoning-precision trade-off, advancing robust, general-purpose robotic manipulation. Project page: https://geniereasoner.github.io/GenieReasoner/
@article{arxiv.2512.24125,
title = {Unified Embodied VLM Reasoning with Robotic Action via Autoregressive Discretized Pre-training},
author = {Yi Liu and Sukai Wang and Dafeng Wei and Xiaowei Cai and Linqing Zhong and Jiange Yang and Guanghui Ren and Jinyu Zhang and Maoqing Yao and Chuankang Li and Xindong He and Liliang Chen and Jianlan Luo},
journal= {arXiv preprint arXiv:2512.24125},
year = {2026}
}