Unverified50% confidenceDecision RuleExact time
عند تبنّي نظام معالجة عربية يُنصح بعدم قبول مؤشر دقة واحد من المورّد وطلب اختبار فعلي على بياناتك بلهجة جمهورك مع تقييم منفصل لكل لهجة
1
Sources
50%
Confidence
Long-term
Relevance
10/11/2026
First Seen
Sources
Related Entities
Related Claims
Unverifiedيُنصح بالتقييم المنفصل لكل لهجة لأن مؤشر الدقة الإجمالي المرتفع قد يُخفي فشلاً ذريعاً في لهجة بعينها76% similarUnverifiedUsing a language model to judge objective factual verification—such as whether a tool call is correct or whether returned data is truthful—is mismatched with the role of language quality evaluation.68% similarUnverifiedSafety alignment training data is most comprehensive in English, and may have significant coverage gaps in languages such as Arabic and Urdu, constituting a multilingual jailbreak vulnerability.68% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/999229API
curl https://kongchang.com/api/v1/knowledge/claims/999229MCP
get_claim(id=999229)