Unverified50% confidenceEventExact time
Developers benchmarked multilingual content moderation with 6 models, 25 languages, and 1940 texts, at a total API call cost of $1.84.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/13/2026
First Seen
Valid until: 1/11/2027
Sources
Related Entities
Related Claims
UnverifiedIn experiments with 85 million non-embedding parameters and matched training token budgets, unconstrained fitting produced higher held-out average bits-per-byte across all 12 languages under a shared unconstrained decoder.64% similarUnverifiedThe final trained model has approximately 400 million (0.4B) parameters, targets Bengali as the primary language, and uses a corpus of approximately 2 billion tokens.63% similarUnverifiedThe research team tested the Functionalizer on six corpora covering both natural language and code text.62% similarUnverified阿里建立了覆盖超过30种语言的高质量预训练语料库，总量据估计超过18万亿token62% similarUnverifiedService providers measure AI resource usage in tokens, with one token corresponding to approximately 4 English characters or 1–2 Chinese characters.61% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1015385API
curl https://kongchang.com/api/v1/knowledge/claims/1015385MCP
get_claim(id=1015385)