Unverified50% confidenceFactExact time
In the selective prediction framework, the utility function is typically written as U = α·P(correct) - β·P(wrong), where α and β represent the reward weight for correct answers and the penalty weight for wrong answers, respectively.
1
Sources
50%
Confidence
Long-term
Relevance
10/12/2026
First Seen
Sources
Related Entities
Related Claims
Unverified策略梯度定理表明期望回报对策略参数的梯度可表示为 ∇J(θ) = E[∇log π_θ(a|s) · A(s,a)]，其中A是优势函数72% similarUnverified自回归模型逐Token从左到右生成，本质是马尔可夫过程，在逻辑推理中存在错误积累效应67% similarUnverifiedThe autoregressive token prediction paradigm is built on correlation rather than causality.64% similarUnverifiedRidge Regression is a regression algorithm that adds an L2 regularization term on top of ordinary linear regression, penalizing model coefficients while minimizing prediction error to prevent overfitting.64% similarUnverifiedForcing the model in an Opus 5.5 prompt to display the full reasoning chain in its response may cause the model to refuse or produce lower-quality output.64% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1002065API
curl https://kongchang.com/api/v1/knowledge/claims/1002065MCP
get_claim(id=1002065)