# Embedding very sensitive to punctuation **URL:** \ **Category:** API\ **Tags:** embeddings, ada, ada002\ **Created:** [December 6, 2023, 1:13pm UTC](https://community.openai.com/t/embedding-very-sensitive-to-punctuation/546205 "2023-12-06T13:13:25Z")\ **Posts on this page:** 5\ **Page:** 1
**Author:** ![leo.bachelier.ext](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/leo.bachelier.ext/32/147246_2.png) [@leo.bachelier.ext](https://community.openai.com/u/leo.bachelier.ext)\ **Post date:** [December 6, 2023, 1:13pm UTC](https://community.openai.com/t/embedding-very-sensitive-to-punctuation/546205/1 "2023-12-06T13:13:25Z")
I have been using OpenAI Embeddings specifically text-embedding-ada-002 and noticed it was very sensitive to punctuation even. I have around 1000 chunks and need to extract each time the 15 most similar chunks to my query. I have been testing my query without punctuation and when I add a dot ‘.’ at the end of my query it changes the initial set I got from the retriever with the query without punctuation (some chunks are the same but new ones may appear or the initial order is different). - Have you noticed anything similar ? - Is it the basic behaviour of this embedding to be that sensitive to punctuation ? - Is there a way to make it more robust to minor changes in the query ? FYI: I am using PGvector to store my chunks vectors ---
**Author:** ![lmccallum](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/lmccallum/32/363_2.png) [@lmccallum](https://community.openai.com/u/lmccallum)\ **Post date:** [December 6, 2023, 8:29pm UTC](https://community.openai.com/t/embedding-very-sensitive-to-punctuation/546205/2 "2023-12-06T20:29:32Z")
I haven’t noticed punctuation, but I have noticed a significant downgrade in performance using ada-002 compared to the davinci-001 embeddings model. I am really frustrated because I re-embedded all my texts, and now the results aren’t fit for my use case. Before, the most relevant text always showed up in the first 1-5 search results, now it’s the 50th search result! ---
**Author:** ![anon10827405](https://avatars.discourse-cdn.com/v4/letter/a/ed8c4c/32.png) [@anon10827405](https://community.openai.com/u/anon10827405)\ **Post date:** [December 6, 2023, 8:32pm UTC](https://community.openai.com/t/embedding-very-sensitive-to-punctuation/546205/3 "2023-12-06T20:32:14Z")
> [@leo.bachelier.ext](#): > > Have you noticed anything similar ? > Is it the basic behaviour of this embedding to be that sensitive to punctuation ? Yes. Grammar do be like that. > [@leo.bachelier.ext](#): > > Is there a way to make it more robust to minor changes in the query ? Have you tried normalizing/correcting the text using GPT and then embedding it? ---
**Author:** ![leo.bachelier.ext](https://sea2.discourse-cdn.com/openai1/user_avatar/community.openai.com/leo.bachelier.ext/32/147246_2.png) [@leo.bachelier.ext](https://community.openai.com/u/leo.bachelier.ext)\ **Post date:** [December 8, 2023, 10:12am UTC](https://community.openai.com/t/embedding-very-sensitive-to-punctuation/546205/4 "2023-12-08T10:12:58Z")
Actually, it’s more in the query I am sending to the retriever, if I add a dot at the end it changes the set returned. Depending on the query sometimes the version with the dot returns what I am looking for and sometimes it’s the version without the dot. So not sure how to normalize it correclty. ---
**Author:** ![anon10827405](https://avatars.discourse-cdn.com/v4/letter/a/ed8c4c/32.png) [@anon10827405](https://community.openai.com/u/anon10827405)\ **Post date:** [December 8, 2023, 12:43pm UTC](https://community.openai.com/t/embedding-very-sensitive-to-punctuation/546205/5 "2023-12-08T12:43:14Z")
What was your chunking strategy? Could you provide some examples of the data you chunked?