Unverified50% confidenceDecision RuleExact time
LiveNerf's criterion is: if Opus 5.5's score deviates by more than 7.5 points, it is treated as a potential signal of capability degradation.
1
Sources
50%
Confidence
Medium-term (~90 days)
Relevance
9/28/2026
First Seen
Valid until: 12/27/2026
Sources
Related Entities
Related Claims
UnverifiedIn max mode, GPT-5.6's score is actually slightly lower than in ultra mode, confirming the overthinking phenomenon.69% similarUnverified在MATH数据集的Level 5竞赛级难题上，前沿模型准确率仍低于60%68% similarUnverifiedFor categorical policies, as the output distribution becomes increasingly sharp, the expected score-gradient norm approaches zero, resulting in the weakest updates for tokens that are confident but incorrect.67% similarUnverifiedThe authors used a DINO loss weight of 0.5 in this experiment and believe this value may be too high, with reducing the weight being the primary direction for next-step optimization.67% similarUnverifiedThe standard margin of error for Gallup polls is typically kept within ±3–4 percentage points at a 95% confidence level.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/955019API
curl https://kongchang.com/api/v1/knowledge/claims/955019MCP
get_claim(id=955019)