Moderate Vector DB Question 96 of 223

How should you update a vector index when documents change?

GenAI / LLM · Speak this in 60–90 seconds · Faridabad & Delhi NCR

Without indexScan every row

With indexJump to keys

CostWrites slower

Re-embed changed chunks, upsert by stable IDs, and delete vectors for removed content.

LLMs use tokens (not full words) because it helps them:

Vector DB questions separate

people who only read docs from people who shipped.

Keep it short, concrete,

and tied to GenAI / LLM work.

Stable token IDs

Each piece maps to a number the network can learn.

Fits the model

Fixed pieces are what transformers expect as input.

Before the model can read a sentence, it goes through these steps:

1
Re-embed changed chunks, upsert
by stable IDs, and delete vectors for removed content.
2
Stale embeddings cause wrong
answers that look authoritative.
3
Use a content hash
so you do not re-embed unchanged bytes.
4
Context mix
Attention looks at nearby tokens together.
5
Next token
The model scores what should come next.
6
Decode
IDs turn back into readable text.

Let's see how a real sentence is tokenized (tokens may vary by model):

Input text

“Stale embeddings cause wrong answers that look authoritative.”

Tokenized output

Staleembeddingscausewronganswersthat

Token IDs (example)

2987408337471632900

Note: Actual tokens and IDs depend on the tokenizer (e.g., GPT, Llama, etc.).

Re-embed changed chunks, upsert by stable IDs, and delete vectors for removed content. Stale embeddings cause wrong answers that look authoritative.