3 hints available

EasyAI EngineeringTheoryRAG Architecture Vector Databases LLM Evaluation

Embedding Similarity — Semantic Reasoning

You have embedded the following 5 sentences using text-embedding-3-small and computed pairwise cosine similarities:

Sentence A: "How do I reset my account password?" Sentence B: "I forgot my login credentials and cannot access my account." Sentence C: "The password reset email is not arriving in my inbox." Sentence D: "How do I cancel my subscription?" Sentence E: "What is the refund policy for cancelled plans?"

Without running the actual embeddings, reason about the cosine similarity scores:

Which two sentences have the highest cosine similarity? Explain why.
Which pair has the lowest cosine similarity? Explain why.
A user searches with Sentence A. Which documents would a RAG system retrieve (top 2)? Would this retrieval be useful?
Why might keyword search fail to identify the relationship between Sentence A and Sentence B?

Sign in to attempt this problem

Free account gives you full access to community problems with the complete solution reveal: golden answer, senior walkthrough, and score breakdown, after submission.

Start free →Already have an account? Sign in