Expired50% confidenceBenchmarkExact time
obliteratus modified 841 out of 850 tensors; 44.8% of HarmBench responses never completed the thinking phase, making it the only variant to show significant degradation on intelligence tests.
1
Sources
50%
Confidence
Short-term (~14 days)
Relevance
9/7/2026
First Seen
Valid until: 9/21/2026(expired)
Sources
Related Entities
Related Claims
UnverifiedA 95% confidence level means that if the same methodology were used to conduct 100 surveys, approximately 95 of them would produce results falling within the margin of error.61% similarUnverifiedAIME has only 30 samples; claiming superiority by bolding a score of 80 against a baseline of 79 after running 4 seeds is statistically untenable.60% similarUnverifiedMuse Glimmer在Artificial Analysis第二版测试中幻觉率高达82%，知识工作和幻觉是其最大弱点59% similarUnverified一个 90% 成功、10% 出错的工具调用系统几乎无法投入实际使用，因为单次错误会污染后续上下文造成级联失败59% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/871878API
curl https://kongchang.com/api/v1/knowledge/claims/871878MCP
get_claim(id=871878)