Unverified50% confidenceFactExact time
The researchers conducted bitext mining tasks between English and more than 200 other languages in cross-lingual scenarios.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/19/2026
First Seen
Valid until: 1/17/2027
Sources
Related Entities
Related Claims
UnverifiedThe researchers used the MTEB and MMTEB benchmarks, covering 18 languages, and conducted more than 200 monolingual evaluations.72% similarUnverifiedThe West Romance language experiment was conducted on four languages—French, Portuguese, Spanish, and Italian—using English as an anchor point and 150 sets of four-language parallel sentences as the data foundation.69% similarUnverifiedThe study tested the introduction of an auxiliary language identification task as an intervention to strengthen language differentiation capability.68% similarUnverifiedWhen the same question is posed in different languages, the documents retrieved by a multilingual dense retriever differ enormously.68% similarUnverifiedBitext mining is a classic evaluation task for cross-lingual embeddings: given sentence sets in each of two languages, the system must find, for each sentence, the semantically corresponding translated sentence in the other language.67% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1064858API
curl https://kongchang.com/api/v1/knowledge/claims/1064858MCP
get_claim(id=1064858)