Unverified50% confidenceFactExact time
MQM classifies, counts, and weights errors by error type (such as accuracy and fluency) and severity (such as minor and major), while ESA requires annotators to precisely circle the text spans corresponding to errors.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDeterministic scorers can reliably catch field format errors, but averaging them across the entire suite cannot tell you where the errors are concentrated68% similarUnverified作者归纳出四类高频确定性工具调用错误：可选字段传null值、数组格式错误、空对象包裹、参数值缺失68% similarUnverifiedIn document parsing checks, the correspondence between cells and headers is most critical; numbers may be correctly recognized yet fall in the wrong row, making this type of structural error the most hidden.64% similarUnverifiedLLMs have a strong tendency toward error tolerance and completion, and when faced with misaligned table data they typically do not report errors but instead produce seemingly reasonable inferences based on the erroneous data.63% similarUnverifiedWhen OCR output is fed directly to an LLM, recognition errors are amplified and propagated, causing the model to generate plausible-sounding but incorrect conclusions based on erroneous text62% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1012333API
curl https://kongchang.com/api/v1/knowledge/claims/1012333MCP
get_claim(id=1012333)