semantic collapse
The phenomenon in which text loses semantic information once separated from its context. It refers to the loss of structural information (such as headings and hierarchy) during text splitting, causing embedding vectors to collapse and semantic coordinates to converge toward a common attractor.
Timeline (last 90 days)
Once text is divorced from its original structural context, its semantics collapse, manifesting as text that should be distinguishable obtaining nearly identical vectors in the embedding space
Semantic collapse occurs before the text is encoded into vectors, so switching to a stronger embedding model usually cannot fix this kind of retrieval bug caused by loss of context
All Facts (2)
Once text is divorced from its original structural context, its semantics collapse, manifesting as text that should be distinguishable obtaining nearly identical vectors in the embedding space
50%UnverifiedSemantic collapse occurs before the text is encoded into vectors, so switching to a stronger embedding model usually cannot fix this kind of retrieval bug caused by loss of context
50%