LLMs achieve good readability on Vietnamese legal texts but frequently err in fine-grained legal accuracy and reasoning, with Incorrect Example and Misinterpretation as top failure modes.
Large legal fictions: Profiling legal hallucinations in large language models
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
From Benchmarking to Reasoning: A Dual-Aspect, Large-Scale Evaluation of LLMs on Vietnamese Legal Text
LLMs achieve good readability on Vietnamese legal texts but frequently err in fine-grained legal accuracy and reasoning, with Incorrect Example and Misinterpretation as top failure modes.