Unverified50% confidenceFactExact time
为模型设定角色标签会激活对应的语料分布，使输出概率分布向该领域专业表达倾斜，从而提升输出质量
1
Sources
50%
Confidence
Long-term
Relevance
7/9/2026
First Seen
Sources
提示词工程入门：原理、方法与实战案例全解析
bilibili大模型零基础入门_6/30/2026
Related Claims
Unverified用户输入的每个词都会成为模型的上下文，改变模型接收到的数学输入，进而影响解码阶段的概率分布和输出内容72% similarUnverifiedSetting a role in a prompt changes the model's 'prior distribution,' influencing the token probability space for subsequent generation.71% similarUnverified上下文偏置（Contextual biasing）在解码阶段介入，通过向模型提供期望词表提高特定词的输出概率，实现方式包括浅层融合、深层偏置和 TCPGen 等架构71% similarUnverifiedExcellent prompt engineering can improve the output quality of the same model several-fold, and in certain tasks can make small models match the performance of large models.70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/403234API
curl https://kongchang.com/api/v1/knowledge/claims/403234MCP
get_claim(id=403234)