Unverified50% confidenceConceptExact time
RAG changes what the model can see, while fine-tuning changes how the model performs.
1
Sources
50%
Confidence
Long-term
Relevance
10/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedRAG and fine-tuning address two different types of problems: one concerns what the model knows, and the other concerns how the model expresses itself and behaves.73% similarUnverifiedThe rows that change are precisely the rows on which models disagree with one another72% similarUnverifiedRAG工作流通常涉及两类模型：嵌入模型（将文本转换为向量用于相似度检索）和对话大模型（综合检索结果生成答案）71% similarUnverifiedThe prompt determines what the model hears, while context determines what the model sees; the two are different concepts.70% similarVerifiedFine-tuning can deeply alter a model's behavioral patterns, whereas prompt engineering can only operate within the boundaries of the model's existing capabilities70% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1020345API
curl https://kongchang.com/api/v1/knowledge/claims/1020345MCP
get_claim(id=1020345)