Unverified50% confidenceTradeoffExact time
Vector databases achieve semantic retrieval by mapping text to high-dimensional vectors and computing cosine similarity, and can recall content that differs literally but is close in meaning; however, their embedding models' training data and parameters are opaque, similarity thresholds are hard to set intuitively, and an Agent cannot infer from the ranking results why a particular memory was selected.
1
Sources
50%
Confidence
Long-term
Relevance
10/15/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedEmbedding models map queries and documents into the same vector space and match them using measures such as cosine similarity; when a query uses domain abbreviations, internal codenames, or wording whose distribution differs greatly from the training corpus, the correct document may be ranked very far down or may not even appear in the candidate set.73% similarUnverifiedHindsight-type memory components typically rely on vector databases, converting events into high-dimensional vectors via embedding models and retrieving them using cosine similarity or approximate nearest-neighbor algorithms.73% similarUnverifiedVector databases cannot express structured relationships (such as the source user, session, and time of a memory), cannot perform precise filtering and transactional updates, and their similarity thresholds also lack interpretability.72% similarUnverified向量数据库的核心优势在于语义级模糊检索，语义相近的内容在高维向量空间中距离更近71% similarUnverifiedVector retrieval performs terribly at looking up exact identifiers (such as contract numbers, customer names, date ranges, and material codes), because embedding models compress text into dense vectors, and similarity ranking has no binding force for exact equality.71% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1037682API
curl https://kongchang.com/api/v1/knowledge/claims/1037682MCP
get_claim(id=1037682)