@inproceedings{sui-etal-2025-grounding,
    title = "From Grounding to Manipulation: Case Studies of Foundation Model Integration in Embodied Robotic Systems",
    author = "Sui, Xiuchao  and
      Tian, Daiying  and
      Sun, Qi  and
      Chen, Ruirui  and
      Choi, Dongkyu  and
      Kwok, Kenneth  and
      Poria, Soujanya",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2025",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/name-variant-enfa-fane/2025.findings-emnlp.69/",
    doi = "10.18653/v1/2025.findings-emnlp.69",
    pages = "1324--1340",
    ISBN = "979-8-89176-335-7",
    abstract = "Foundation models (FMs) are increasingly applied to bridge language and action in embodied agents, yet the operational characteristics of different integration strategies remain under-explored{---}especially for complex instruction following and versatile action generation in changing environments. We investigate three paradigms for robotic systems: end-to-end vision-language-action models (VLAs) that implicitly unify perception and planning, and modular pipelines using either vision-language models (VLMs) or multimodal large language models (MLLMs). Two case studies frame the comparison: instruction grounding, which probs fine-grained language understanding and cross-modal disambiguation; and object manipulation, which targets skill transfer via VLA finetuning. Our experiments reveal trade-offs in system scale, generalization and data efficiency. These findings indicate design lessons for language-driven physical agents and point to challenges and opportunities for FM-powered robotics in real-world conditions."
}