Unverified50% confidenceFactExact time
Character Error Rate (CER) uses the same calculation logic as WER but operates at the character level, making it more suitable for languages such as Chinese and Japanese that lack clear word boundaries.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedCER is the naturally preferred metric for languages without explicit word boundaries, such as Chinese and Japanese.74% similarUnverified中文、日文等表意文字因BPE词表以英语语料为主，往往需要更多Token表达相同信息量67% similarUnverifiedUTF-8对中文字符采用3字节编码，范围为U+4E00至U+9FFF的CJK统一汉字65% similarUnverified不同分词算法（BPE、WordPiece等）导致同一段中文文本在不同模型中消耗的Token数量可能相差20%-40%65% similarUnverified新一代分词器通常能以更少的Token表达相同语义信息，对代码及中文、日文等非拉丁语言处理效率更高64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/895553API
curl https://kongchang.com/api/v1/knowledge/claims/895553MCP
get_claim(id=895553)