Unverified50% confidenceFactExact time
Temperature scaling is a calibration method that attempts to calibrate logits into more reliable probability estimates.
1
Sources
50%
Confidence
Long-term
Relevance
9/12/2026
First Seen
Sources
Related Entities
Related Claims
VerifiedThe large model temperature parameter adjusts the flatness of the output probability distribution by dividing logits by the temperature value before applying softmax normalization.79% similarUnverified置信度校准的解决方案包括temperature scaling（温度缩放）、verbalized confidence（显式输出置信度分数）、consistency-based estimation（多次采样检查一致性）75% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/912213API
curl https://kongchang.com/api/v1/knowledge/claims/912213MCP
get_claim(id=912213)