Metraipolis › archive › dc_46b94ed079c2
EvoGenUI-Bench: Evaluating LLMs as Multi-Turn Generative UI Assistants — arXiv:2608.29387
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.29387
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T07:44:19.392Z
- SHA-256 of the archived text
321fa6e53808ad421484670dd6e13ad4c0523a75edb8d28cf90ebe37bb782b87- Stable URL of this page
https://metraipolis.com/documento/dc_46b94ed079c2
How to cite
EvoGenUI-Bench: Evaluating LLMs as Multi-Turn Generative UI Assistants — arXiv:2608.29387 — arXiv (en), https://arxiv.org/abs/2608.29387 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_46b94ed079c2 (Metraipolis archive dc_46b94ed079c2, sha256 321fa6e53808ad421484670dd6e13ad4c0523a75edb8d28cf90ebe37bb782b87)
BibTeX
@misc{dc46b94ed079c2,
title = {EvoGenUI-Bench: Evaluating LLMs as Multi-Turn Generative UI Assistants — arXiv:2608.29387},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.29387},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_46b94ed079c2; sha256 321fa6e53808ad421484670dd6e13ad4c0523a75edb8d28cf90ebe37bb782b87},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.29387. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Large language models can generate interactive web interfaces, but reliable generative UI requires maintaining an executable artifact as user requests evolve. We introduce EvoGenUI-Bench, 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. We execute generated artifacts in a browser and evaluate them using screenshots, source and DOM evidence, actor traces, and runtime logs. Beyond turn-level and episode-level success, we measure cross-turn retention with Adjacent Pass Retention. Across eight models, even the strongest achieves 74.9% Turn Pass while completing only 37.3% of five-turn episodes; APR further falls to 52.4% on tool-grounded tasks. Diagnostic analysis shows that presentation failures center on information architecture, interaction failures on derived-state propagation and affordance binding, and tool-grounded failures additionally involve external-state grounding and requirement decomposition. These results reframe generative UI evaluation from judging isolated outputs to testing whether interface behavior, derived state, external state, and assistant claims remain synchronized as the artifact evolves. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.AI). Categorie arXiv: Artificial Intelligence (cs.AI) ; Software Engineering (cs.SE). Pagina del lavoro: https://arxiv.org/abs/2608.29387 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T07:28:26.552Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).