Unverified50% confidenceFactExact time
Multilingual large language models often give inconsistent answers to semantically identical questions in different languages, a phenomenon known as the Cross-Lingual Consistency (CLC) deficiency.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCross-lingual transfer in multilingual models mainly relies on a shared multilingual representation space, which exhibits a significant alignment gap between language pairs with large differences in grammatical structure.74% similarUnverified跨语言词义消歧（CLWSD）不同于单语词义消歧，需要在源语言多义词与目标语言多个可能翻译之间做选择74% similarUnverified通用大模型在特定语言的方言处理、形态变化丰富的语言以及文化特定表达上仍存在明显不足72% similarUnverifiedLarge language models have a miscalibration problem, often answering in a confident tone regardless of whether the answer is correct or incorrect.72% similarUnverifiedCross-lingual jailbreaking exploits the asymmetry between a model's concept space and its language space, as the learning of safety-refusal behavior is highly language-specific.72% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/902516API
curl https://kongchang.com/api/v1/knowledge/claims/902516MCP
get_claim(id=902516)