AI RESEARCH
The Bitter Lesson of Diffusion Language Models for Agentic Workflows: A Comprehensive Reality Check
arXiv CS.CL
•
ArXi:2601.12979v3 Announce Type: replace 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. 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