Unverified50% confidenceBenchmarkExact time
The eight evaluated models generally have lower accuracy in Bengali than in English, and among them Mistral-Nemo has the smallest cross-lingual accuracy gap, at 5.12 percentage points.
1
Sources
50%
Confidence
Long-term
Relevance
11/10/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedLEACE fitted on English removed less than one percentage point of Bengali metaphor category accuracy, whereas LEACE fitted on Bengali removed 4-11 percentage points.74% similarUnverifiedExperiments find that LLM answer quality and coverage in certain languages are significantly lower than in others, causing cross-lingual information inequality.70% similarUnverifiedThe author emphasizes that 5.05% is a perplexity reduction at the language model validation level, not an improvement in benchmark accuracy.69% similarUnverifiedيُنصح بالتقييم المنفصل لكل لهجة لأن مؤشر الدقة الإجمالي المرتفع قد يُخفي فشلاً ذريعاً في لهجة بعينها68% similarUnverifiedThe study shows that certain models exhibit significant performance degradation on distorted statements in East Asian languages (Chinese, Japanese, and Korean), with cross-lingual performance gaps of up to 28 percentage points and relative declines as high as 49%.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1154785API
curl https://kongchang.com/api/v1/knowledge/claims/1154785MCP
get_claim(id=1154785)