Unverified50% confidenceSolutionExact time
The Plan-then-Retrieve (PtR) paradigm lets the LLM first generate several abstract tool stubs (Abstract Tool Stubs) based on user intent as retrieval queries, then match real tools in the registry accordingly.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedIn LLM applications, a single user question typically corresponds to one trace, and the internal model calls, vector retrievals, and tool executions are each a child span.72% similarUnverifiedLLM 与形式化证明系统结合主要有生成式（Generative）和搜索引导式（Search-Guided）两种范式67% similarUnverifiedTool Retrieval uses vector search to first filter out the 3–5 most relevant tools before providing them to the model.66% similarUnverifiedMulti-query retrieval uses an LLM to automatically generate 3–5 semantically equivalent but differently phrased variant queries for the same question, retrieves results for each separately, and takes the union to improve recall.63% similarUnverifiedReranking uses a Cross-Encoder model to score candidate documents retrieved in the initial retrieval pairwise, then fine-ranks the Top-K to feed into the LLM.63% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1040080API
curl https://kongchang.com/api/v1/knowledge/claims/1040080MCP
get_claim(id=1040080)