Two sentence embeddings have a high cosine similarity. What does this suggest?
Why B is the answer
Similar meaning
Embedding similarity reflects meaning, not length, authorship or storage.
In detail
- High cosine similarity suggests similar meaning.
- Cosine compares direction, not length.
Use case
An illustrative scenario, not a real incident.
"How do I reset my password" and "I forgot my login" have a high cosine similarity, though they share few words.
See it in different ways
flowchart LR A["Text"] --> B["Embedding model"] B --> C["Vector"] C --> D["Vector database"] E["Query vector"] --> D D --> F["Nearest passages"]
sequenceDiagram participant U as Query participant E as Embedding model participant V as Vector database U->>E: Text E-->>U: Vector U->>V: Find the nearest V-->>U: Similar passages
flowchart TD
Q{"How close are the vectors?"}
Q -->|"High cosine similarity"| A0["Similar meaning"]
Q -->|"Low cosine similarity"| A1["Different meaning"]Key point: High cosine similarity suggests similar meaning.
Why the other options are wrong
- A. They are stored close together on the disk
- Disk location is unrelated.
- C. They hold exactly the same number of tokens
- Token count is unrelated.
- D. They were written by the same person
- Authorship is unrelated.