Unverified50% confidenceFactExact time
When a bilingual model's training L1 matches the reader's native language, its predictions of L2 reading times are more accurate than the monolingual baseline and the mismatched group.
1
Sources
50%
Confidence
Long-term
Relevance
10/23/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe study design included three types of predictors: the matched group (model training L1 consistent with the reader's native language), the monolingual baseline (a pure-English monolingual model), and the mismatched group (model training L1 inconsistent with the reader's native language).73% similarVerifiedLLMs acquire language capabilities by performing unsupervised pre-training on massive text corpora (predicting the next token).72% similarUnverifiedPerez et al.的《Discovering Language Model Behaviors with Model-Written Evaluations》证实经过RLHF训练的模型普遍存在谄媚倾向71% similarUnverifiedThe surprisal of monolingual language models has been repeatedly validated as an effective predictor of native speakers' reading times.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1088199API
curl https://kongchang.com/api/v1/knowledge/claims/1088199MCP
get_claim(id=1088199)