A multi-agent, adaptive benchmark framework for evaluating LLM-generated frontend projects, with a 100-query benchmark, an 11-model leaderboard, and 86% agreement with human UI ratings.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
LiveEvalBench: Toward Open-World Evaluation for Web Generation
A multi-agent, adaptive benchmark framework for evaluating LLM-generated frontend projects, with a 100-query benchmark, an 11-model leaderboard, and 86% agreement with human UI ratings.