Unverified50% confidenceFactExact time
LLMs have an extremely low proportion of pre-training data in low-resource languages, which directly affects the model's language understanding ability and prompt response behavior.
1
Sources
50%
Confidence
Long-term
Relevance
11/3/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedLLMs acquire language capabilities by performing unsupervised pre-training on massive text corpora (predicting the next token).78% similarVerifiedThe safety alignment of many models degrades significantly in non-English languages, and low-resource languages are especially easy to exploit because the human annotation data for RLHF alignment training is highly concentrated in English.76% similarUnverifiedThe training data for large language models (LLMs) is mainly human language; if language itself carries hidden biases, then model outputs may also amplify these biases.76% similarUnverifiedIn the NLP field, low-resource languages refer to languages lacking large-scale annotated corpora, pre-trained models, and benchmark datasets, and Nepali falls into this category.76% similarUnverifiedTraditional n-gram models and neural language models trained on a single language assign extremely low probability scores to cross-lingual word sequences, causing the decoder to tend to forcibly map foreign-language words to phonetically similar words in the target language.75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1129067API
curl https://kongchang.com/api/v1/knowledge/claims/1129067MCP
get_claim(id=1129067)