Unverified50% confidenceTradeoffExact time
Agentic Search 效果高度依赖底层模型推理质量，在推理能力弱的模型上可能适得其反
1
Sources
50%
Confidence
Long-term
Relevance
7/13/2026
First Seen
Sources
Claude Code为何放弃RAG，选择Grep搜索代码
Related Claims
Unverified自动化研究智能体与传统超参数搜索或神经架构搜索不同，能够提出全新的模型结构假设而非仅在预定义搜索空间中优化73% similarUnverifiedAgent只应在无法用if-else或规则引擎覆盖的真正推理需求（如理解模糊意图、权衡不完整信息源）时介入71% similarUnverifiedThe criteria by which model auto-routing algorithms judge the 'optimal model' often diverge from users' subjective preferences, and when the algorithm prioritizes cost or latency it may sacrifice output quality.70% similarUnverifiedAgentic Search 在中小规模、结构清晰的代码库表现更好，超大规模复杂场景尚无定论69% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/502768API
curl https://kongchang.com/api/v1/knowledge/claims/502768MCP
get_claim(id=502768)