EvoGenUI-Bench: Evaluating LLMs as Multi-Turn Generative UI Assistants
EvoGenUI-Bench is introduced, a benchmark for multi-turn interface maintenance comprising 150 five-turn tasks and 750 turns across three scenarios: information presentation, executable interaction, and tool-grounded external state.
Yue Peng, Lan-Ke Xia, Zi-Han Wang et al.
· 1 citation