Unverified50% confidenceBenchmarkExact time
At a performance degradation cost of 7.6%, DarKnight recovered 48.2% of baseline failure cases.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
10/15/2026
First Seen
Valid until: 1/13/2027
Sources
Related Entities
Related Claims
UnverifiedDarKnight's 83.2% pass rate with an open-weight VLM exceeds Claude Opus 5's 44% on the same task.60% similarUnverified测评者在Minecraft复刻测试中给Sonnet 5评为65分（满分100），存在方块破碎无动画、无库存系统等缺陷53% similarUnverifiedIn a bug-fix test case, DeepSeek located and modified the calculation module so that all 8 tests passed, correcting the report from an erroneous $108.06 to the correct $65.05, taking approximately 1 minute and 20 seconds under max thinking.53% similarUnverified根据基于Steam成就解锁率的统计研究，大多数RPG游戏的最终章节成就解锁率仅有15%-30%53% similarUnverifiedIn the 30-day holdout evaluation, TabPFN v2 had the lowest error with an MAE of 10.3 and RMSE of 14.6, and the 95% confidence interval for its MAE advantage over B2 was −17.0 to −6.9.52% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1038923API
curl https://kongchang.com/api/v1/knowledge/claims/1038923MCP
get_claim(id=1038923)