REAL is a 5,316-question Chinese real estate benchmark for LLMs covering memory, comprehension, reasoning, and hallucination, and it shows current models score only about 50% on average.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.AI 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
REAL: Benchmarking Abilities of Large Language Models for Housing Transactions and Services
REAL is a 5,316-question Chinese real estate benchmark for LLMs covering memory, comprehension, reasoning, and hallucination, and it shows current models score only about 50% on average.