Unverified50% confidenceBenchmarkExact time
In frozen replay test scenarios, to achieve complete recall, some cases require reading 290 to 426 out of 407 to 454 candidates.
1
Sources
50%
Confidence
Long-term
Relevance
9/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedDiscoBench最终生成211个测试样本，包含463个歧义实例64% similarUnverifiedIn comprehensive benchmark tests, Sol scores 55 (approximately 78.6%), Terra scores 44 (approximately 62.9%), and Luna scores 31 (approximately 44.3%).56% similarUnverifiedThis test generated a total of 147 model calls.55% similarUnverifiedIn tests covering context lengths of 512–4096, increasing the draft length from 4 tokens to 16 tokens resulted in only a 0.2%–1.3% increase in overall inference latency.52% similarUnverified叠加cv=5的5折交叉验证后，144种参数组合需训练144×5=720次模型51% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/922247API
curl https://kongchang.com/api/v1/knowledge/claims/922247MCP
get_claim(id=922247)