Unverified50% confidenceFactExact time
Mixed Error Rate (MER) incorporates formatting differences such as punctuation and capitalization into the statistics on top of WER, but likewise cannot distinguish keywords from non-keywords.
1
Sources
50%
Confidence
Long-term
Relevance
11/10/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedWord Error Rate (WER) is calculated by dividing the sum of substitution, deletion, and insertion errors between the recognition output and the reference text by the total number of words in the reference text.67% similarUnverifiedWER is calculated as (number of substituted words + number of deleted words + number of inserted words) / total number of reference words, and is extremely sensitive to text formatting (case, punctuation, abbreviations, spelling variants), so normalization is usually required before evaluation.66% similarUnverifiedIn the empirical tests, the MiniLM Cross-Encoder, biased toward lexical matching and sensitive to the scrambled text sequences produced by PDF parsing, pushed both correct answer chunks out of the Top 5, causing ranking degradation.65% similarUnverifiedDense embeddings excel at conceptual clustering, but perform poorly on exact lexical matching (such as serial numbers, error codes, domain terms, and abbreviations).64% similarUnverifiedCharacter Error Rate (CER) is defined as the edit distance (insertions + deletions + substitutions) divided by the total number of characters in the reference text; Word Error Rate (WER) is analogous but operates on words.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1153735API
curl https://kongchang.com/api/v1/knowledge/claims/1153735MCP
get_claim(id=1153735)