Get Started
Research questionHow can we reliably evaluate specialised style in LLM-generated Chinese legal text?LLM-generated legal text may be factually sound yet violate implicit conventions of legal writing. Human assessment is expensive, while reference-based metrics and LLM judges may conflate content with style or produce opaque, inconsistent judgments.
AI
Evaluation & Benchmarks
Natural Language Processing
Latest papersRecent research connected to this question, newest first.CLASE: A Hybrid Method for Chinese Legalese Stylistic EvaluationThe evidence concerns Chinese legal documents and reports evaluation on 200 documents against human judgments, including alignment and interpretable score breakdowns. The source describes a reference-free hybrid using linguistic features and experience-guided LLM judging learned from authentic and LLM-restored contrastive document pairs.research paper · Sep 2, 2026
Related questions
How can we tell whether generated text matches an author's style across pragmatic contexts?How can we tell whether agreement among LLM judges reflects human alignment or shared blind spots?How can authorship attribution remain reliable for long-form LLM text across languages and distribution shifts?How can legal AI generate and classify disputed legal issues in Chinese civil litigation?
Home
Topics
Search
Library