A new Chinese government-domain benchmark uses a ModernBERT model trained on subjectively assigned, Beta-diffused relevance labels to score LLM responses.
hard” and “soft
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
-
GovRelBench:A Benchmark for Government Domain Relevance
A new Chinese government-domain benchmark uses a ModernBERT model trained on subjectively assigned, Beta-diffused relevance labels to score LLM responses.