Similarity search vs MMR retrieval in LangChain - Visual Side-by-Side Comparison

Concept Flow - Similarity search vs MMR retrieval

Input Query

↓

Compute Embedding

↓

Find Top K Similar

↓

Return Top K Results

↓

Balance Relevance & Diversity

↓

Return Diverse Results

The flow starts with an input query. Similarity search finds the top similar items directly. MMR retrieval first finds candidates then selects results balancing relevance and diversity.

Execution Sample

LangChain

query = "What is AI?"
results_sim = similarity_search(query, k=3)
results_mmr = mmr_retrieval(query, k=3, lambda_param=0.5)
print(results_sim)
print(results_mmr)

This code runs similarity search and MMR retrieval on the same query, showing different result sets.

Execution Table

Step	Method	Action	Intermediate Result	Output
1	Both	Receive query 'What is AI?'	Query embedding computed	None
2	Similarity Search	Find top 3 most similar documents by embedding similarity	Top 3 docs by similarity scores	Docs A, B, C
3	MMR Retrieval	Find candidate documents (e.g., top 10 by similarity)	Candidate set of 10 docs	Docs A-J
4	MMR Retrieval	Iteratively select docs balancing relevance and diversity using lambda=0.5	Selected docs after each iteration	Docs A, D, F
5	Both	Return final results	Similarity Search returns top 3 similar	Docs A, B, C
6	Both	Return final results	MMR returns diverse top 3 balancing similarity and novelty	Docs A, D, F
7	Both	End	No more steps	Process complete

💡 Both methods finish after selecting k=3 documents; similarity search returns closest only, MMR balances diversity.

Variable Tracker

Variable	Start	After Step 2	After Step 4	Final
query	"What is AI?"	"What is AI?"	"What is AI?"	"What is AI?"
similarity_search_results	[]	[Doc A, Doc B, Doc C]	[Doc A, Doc B, Doc C]	[Doc A, Doc B, Doc C]
mmr_candidates	[]	N/A	[Doc A-J]	[Doc A-J]
mmr_selected	[]	N/A	[Doc A, Doc D, Doc F]	[Doc A, Doc D, Doc F]

Key Moments - 3 Insights

Why does MMR retrieval select different documents than similarity search?

MMR retrieval balances relevance and diversity by iteratively selecting documents that are both similar to the query and different from already selected ones, as shown in steps 3 and 4 of the execution_table.

Does similarity search consider diversity in results?

No, similarity search only ranks documents by similarity score and returns the top k, so results may be very similar to each other, as seen in step 2.

What role does the lambda parameter play in MMR retrieval?

Lambda controls the trade-off between relevance and diversity during selection in MMR retrieval, affecting which documents are chosen in step 4.

Visual Quiz - 3 Questions

Look at the execution_table at step 2, what documents does similarity search return?

A Docs A, D, F

B Docs A, B, C

C Docs A-J

D Docs D, E, F

💡 Hint

Check the 'Output' column for step 2 in the execution_table.

At which step does MMR retrieval balance relevance and diversity?

A Step 4

B Step 3

C Step 2

D Step 5

💡 Hint

Look at the 'Action' column for MMR retrieval in the execution_table.

If lambda in MMR retrieval is set to 1, what would happen to the results?

A No documents are selected

B Results become more diverse

C Results become identical to similarity search

D Results become random

💡 Hint

Lambda=1 means full weight on relevance; see key_moments about lambda's role.

Concept Snapshot

Similarity Search:
- Finds top k documents by embedding similarity
- Results may be similar to each other

MMR Retrieval:
- Finds candidates then selects k balancing relevance and diversity
- Uses lambda to control trade-off

Use similarity search for pure relevance
Use MMR to get diverse relevant results

Full Transcript

This visual execution compares similarity search and MMR retrieval in langchain. Both start by embedding the input query. Similarity search directly finds the top k most similar documents and returns them. MMR retrieval first finds a larger candidate set, then iteratively selects documents balancing relevance to the query and diversity from already selected documents, controlled by a lambda parameter. The execution table shows step-by-step actions and outputs for both methods. Variable tracking shows how results evolve. Key moments clarify why MMR returns different documents and the role of lambda. The quiz tests understanding of these steps and concepts. The snapshot summarizes when to use each method.

Practice

1. What is the main difference between Similarity search and MMR retrieval in Langchain?

easy

A. Similarity search uses machine learning; MMR retrieval uses rule-based filtering.

B. Similarity search finds closest matches; MMR retrieval balances closeness and diversity.

C. Similarity search always returns random results; MMR retrieval returns exact matches.

D. Similarity search is slower than MMR retrieval in all cases.

Solution

1. Step 1: Understand Similarity Search

   Similarity search finds the closest matches to the query based on similarity scores.

2. Step 2: Understand MMR Retrieval

   MMR retrieval balances between closeness and diversity to avoid repeated or very similar results.

3. Final Answer:

   Similarity search finds closest matches; MMR retrieval balances closeness and diversity. -> Option B

4. Quick Check:

   Difference = Closest vs Balanced [OK]

Hint: Similarity = closeness; MMR = closeness plus diversity [OK]

Common Mistakes:

• Thinking MMR returns only closest matches
• Confusing MMR with random retrieval
• Assuming similarity search ensures diverse results

2. Which of the following is the correct way to initialize an MMR retriever in Langchain?

easy

A. mmr_retriever = MMRRetriever(documents, k=5, lambda_mult=0.7)

B. mmr_retriever = MMRRetriever(documents, top_k=5, lambda=0.7)

C. mmr_retriever = MMRRetriever(documents, k=5, lambda=0.7)

D. mmr_retriever = MMRRetriever(documents, top_k=5, lambda_mult=0.7)

Solution

1. Step 1: Check parameter names for MMRRetriever

   The correct parameters include k for number of results and lambda_mult for diversity weight.

2. Step 2: Identify correct parameter usage

   Options B, C, and D use top_k or lambda which are incorrect parameter names.

3. Final Answer:

   mmr_retriever = MMRRetriever(documents, k=5, lambda_mult=0.7) -> Option A

4. Quick Check:

   Correct params = k and lambda_mult [OK]

Hint: MMR uses k and lambda_mult parameters [OK]

Common Mistakes:

• Using 'top_k' instead of 'k'
• Using 'lambda' instead of 'lambda_mult'
• Mixing parameter names from similarity search

3. Given the following code snippet using Langchain's similarity search and MMR retrieval, what will be the main difference in their outputs?

docs = ["apple pie recipe", "apple tart recipe", "banana bread recipe", "cherry pie recipe"]
query = "apple dessert"
sim_results = similarity_search(docs, query, k=2)
mmr_results = mmr_retrieval(docs, query, k=2, lambda_mult=0.5)

medium

A. sim_results will return the two most similar apple recipes; mmr_results will return one apple and one different dessert recipe.

B. sim_results will return random recipes; mmr_results will return the two closest apple recipes.

C. Both sim_results and mmr_results will return the same two apple recipes.

D. sim_results will return only one recipe; mmr_results will return three recipes.

Solution

1. Step 1: Analyze similarity_search output

   Similarity search returns the top 2 closest matches, likely both apple recipes due to query "apple dessert".

2. Step 2: Analyze mmr_retrieval output

   MMR balances similarity and diversity, so it will likely return one apple recipe and one different dessert (like cherry pie) to increase variety.

3. Final Answer:

   sim_results will return the two most similar apple recipes; mmr_results will return one apple and one different dessert recipe. -> Option A

4. Quick Check:

   Similarity = closest; MMR = closest + diverse [OK]

Hint: Similarity = closest matches; MMR adds variety [OK]

Common Mistakes:

• Assuming both return identical results
• Thinking MMR returns random results
• Confusing number of results returned

4. You wrote this code to perform MMR retrieval but get an error:

mmr = MMRRetriever(documents, k=3, lambda_mult=0.6)
results = mmr.get_relevant_documents(query)

What is the likely cause of the error?

medium

A. Calling 'get_relevant_documents' instead of 'retrieve_documents'.

B. Using 'k' instead of 'top_k' when initializing MMRRetriever.

C. Missing import statement for MMRRetriever.

D. Using 'lambda_mult' instead of 'lambda' parameter.

Solution

1. Step 1: Check parameter names in MMRRetriever initialization

   The parameters k and lambda_mult are correct.

2. Step 2: Confirm method name correctness

   get_relevant_documents is a valid method for retrievers.

3. Final Answer:

   Missing import statement for MMRRetriever. -> Option C

4. Quick Check:

   Missing import = NameError [OK]

Hint: Import MMRRetriever before use [OK]

Common Mistakes:

• Forgetting import statement for MMRRetriever
• Confusing method names
• Wrong parameter names like top_k or lambda

5. You want to retrieve 3 documents related to "climate change" using Langchain. You want results that are relevant but also cover different aspects like causes, effects, and solutions. Which retrieval method and parameters should you choose?

hard

A. Use MMR retrieval with k=3 and lambda_mult=0 to get only closest matches.

B. Use similarity search with k=3 for closest matches only.

C. Use similarity search with k=1 repeated 3 times to get variety.

D. Use MMR retrieval with k=3 and lambda_mult=0.8 to balance relevance and diversity.

Solution

1. Step 1: Identify need for relevance and diversity

   You want documents that are relevant but cover different aspects, so diversity is important.

2. Step 2: Choose retrieval method and parameters

   MMR retrieval balances relevance and diversity using lambda_mult. A value like 0.8 prioritizes relevance while introducing some diversity.

3. Final Answer:

   Use MMR retrieval with k=3 and lambda_mult=0.8 to balance relevance and diversity. -> Option D

4. Quick Check:

   Diversity + relevance = MMR with 0 < lambda_mult < 1 [OK]

Hint: Lower lambda_mult increases diversity in MMR [OK]

Common Mistakes:

• Using similarity search only for diverse results
• Setting lambda_mult to 0 disables relevance
• Repeating similarity search calls for variety