Unverified50% confidenceBenchmarkExact time
Under the seen configurations, Dex-One2Many surpasses the baseline method by 6.5%.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedAnthropic's internal testing showed that Fable 5.1 as an advisor paired with Opus 5.5 executing the main task improved performance by 1.7 percentage points compared with using Opus 5.5 alone, but increased cost by 2.1 times.60% similarUnverifiedUnder the same concise prompt and reference-overlap scorer, the published 406M specialized model leads at least five commercially hosted large models by at least 6.2 points on ROUGE-1.60% similarUnverifiedA single top-5 discovery interaction measured in the paper uses only 475 tokens, whereas the same operation under a full-catalog accounting approach requires approximately 42,450 tokens.60% similarUnverifiedLeNet-5约有7层，参数量小58% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1012855API
curl https://kongchang.com/api/v1/knowledge/claims/1012855MCP
get_claim(id=1012855)