Unverified50% confidenceFactExact time
Bayesian decision theory does not directly predict labels; instead, given posterior probabilities, it computes the expected loss for each possible action and then selects the action with the minimum loss.
1
Sources
50%
Confidence
Long-term
Relevance
9/11/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedThe choice of loss function essentially defines the optimization objective of a model, and a wrong loss function means the model is working hard to do the wrong thing.72% similarUnverified将概念推理量化为单一指数可能丢失多维度、情境依赖的推理能力信息，误导下游应用选型决策69% similarUnverifiedThe model's tool-triggering decision is essentially a probabilistic threshold problem: when confidence exceeds the threshold the model tends to answer directly, and only triggers retrieval when confidence falls below it.69% similarUnverified医疗AI中的选择性预测机制在低置信度时拒绝答案转介人工审核，被证明比强制输出答案更能降低临床风险66% similarUnverifiedLonger inference time does not always yield better results; models may fall into the trap of overthinking, and what truly determines output quality is the model's foundational capabilities and degree of alignment.66% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/899634API
curl https://kongchang.com/api/v1/knowledge/claims/899634MCP
get_claim(id=899634)