@inproceedings{lu-etal-2026-bitter,
    title = "The Bitter Lesson of Diffusion Language Models for Agentic Workflows: A Comprehensive Reality Check",
    author = "Lu, Qingyu  and
      Ding, Liang  and
      Zhang, Kanjian  and
      Zhang, Jinxia  and
      Tao, Dacheng",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/declare-journal/2026.acl-long.2036/",
    doi = "10.18653/v1/2026.acl-long.2036",
    pages = "43997--44020",
    ISBN = "979-8-89176-390-6",
    abstract = "The pursuit of real-time agentic interaction has driven interest in Diffusion-based Large Language Models (dLLMs) as alternatives to auto-regressive backbones, promising to break the sequential latency bottleneck. \textbf{However, does such efficiency gains translate into effective agentic behavior?} In this work, we present a comprehensive evaluation of dLLMs (e.g., LLaDA, Dream) across two distinct agentic paradigms: Embodied Agents (requiring long-horizon planning) and Tool-Calling Agents (requiring precise formatting).Contrary to the efficiency hype, our results on Agentboard and BFCL reveal a ``\textbf{bitter lesson}'': current dLLMs fail to serve as reliable agentic backbones, frequently leading to systematically failure. \textbf{(1) In Embodied settings}, dLLMs suffer repeated attempts, failing to branch under temporal feedback. \textbf{(2) In Tool-Calling settings}, dLLMs fail to maintain symbolic precision (e.g. strict JSON schemas) under diffusion noise. To assess the potential of dLLMs in agentic workflows, we introduce \textbf{DiffuAgent}, a multi-agent evaluation framework that integrates dLLMs as plug-and-play cognitive cores. Our analysis shows that dLLMs are effective in non-causal roles (e.g., memory summarization and tool selection) but require the incorporation of causal, precise, and logically grounded reasoning mechanisms into the denoising process to be viable for agentic tasks."
}