Why vector search is not enough and we need BM25

Поділитися
Вставка
  • Опубліковано 16 лис 2024

КОМЕНТАРІ • 60

  • @endre777
    @endre777 Місяць тому +14

    Thanks for the explanation, was super clear.
    We just planning to move from vector search to hybrid, and your explanation on BM25 helps a lot to understand what edge cases it can solve. Appreciate a lot!
    Guess we will see a surge on BM25 due to Anthropic Contextual retrieval paper .

  • @ChocolateMilkCultLeader
    @ChocolateMilkCultLeader Місяць тому +6

    Great video. This is why when building a search engine- I like to use BM25 for sparse search, and use Vector based search later, once most of the corpus has been filtered out. This allows me to stay precise and efficient.
    One additional thing- people often assume that you need a Vector Db for vector search, but you can do completely without. Just store the vectors in a normal DB.

    • @notsojharedtroll23
      @notsojharedtroll23 Місяць тому

      I mean, at the end of the day, the embedsings are data period

    • @stxnw
      @stxnw Місяць тому

      It should be the other way around. Most prompts may not have exact matches. Use vector search first, then BM25 and rerank the results.

  • @shiholololo1053
    @shiholololo1053 День тому

    Waiting for the next videp. I enjoyed the format.

  • @oncedidactic
    @oncedidactic Місяць тому +4

    Nice discussion, thanks! I wish there was more structure to the video so the “why” of the title I served as a main dish, ie let’s define the terms up front, explain how each works, then do why discussion and give a teaser for hybrid approach discussion. Instead there are some gaps and jumps around, which leaves it feeling incomplete or maybe not quite capturing the essence? I have a feeling this is partly a result of editing many clips, so don’t take this feedback too seriously. Cheers

  • @microburn
    @microburn Місяць тому +4

    Nice video. I’ve been on the opposite side of the coin, but I like hearing the balanced argument to keep me educated

  • @matveyshishov
    @matveyshishov Місяць тому

    Thanks, guys, YT recommended me this video, a very pleasant snippet of explanation.
    Trying to work through your website to understand what the service is.

  • @jameswigglesworth8132
    @jameswigglesworth8132 Місяць тому

    Thank you for delving into this important topic!

  • @NicolasEmbleton
    @NicolasEmbleton Місяць тому

    Wonderful explanation. Thank you.

  • @dougunderwood569
    @dougunderwood569 Місяць тому

    Great overview, thank you!

  • @aproperhooligan5950
    @aproperhooligan5950 Місяць тому

    Excellent presentation/explanation. Very useful. Thank you!

  • @andydataguy
    @andydataguy Місяць тому

    Great video! This is one of the most misunderstood concepts. Will def share this next time it comes up!

  • @marka5215
    @marka5215 Місяць тому

    Great explanation. Thank you so much!

  • @weirdsciencetv4999
    @weirdsciencetv4999 Місяць тому +1

    Oh man you are amazing!!
    Love channel I subscribed. Please do a video on working with such graphs using a vector database

  • @roopad8742
    @roopad8742 Місяць тому

    This is so easy to understand, thank you!

  • @badashphilosophy9533
    @badashphilosophy9533 Місяць тому

    this is an amazing explanation. im an instant follower

  • @ashraf_isb
    @ashraf_isb Місяць тому

    thats insightful, thank you so much boss

  • @andrewwalker8985
    @andrewwalker8985 Місяць тому +1

    Why don’t we include semantic dimensions in vectors

  • @amortalbeing
    @amortalbeing Місяць тому

    thanks this was great!

  • @MLGJuggernautgaming
    @MLGJuggernautgaming Місяць тому

    I believe a vector search is still better for rag applications. Bm25 is better for more literal matches. Also what does this have to do with LLMs doing math?

  • @MathsSciencePhilosophy
    @MathsSciencePhilosophy Місяць тому

    The mathematics behind chatGPT is amazing

  • @broccoli322
    @broccoli322 Місяць тому

    Thanks for the video.

  • @theepicosityofpizza
    @theepicosityofpizza Місяць тому +13

    BM25 doesn't do anything to address any of the issues you bring up at the beginning of the video. TF IDF is dumber than vector search in every aspect. It's just much cheaper to run. Not saying it doesn't have value as part of the toolkit but not sure why you spend the first half setting all thes problems with vector search up as if BM25 addresses any of them.

    • @stxnw
      @stxnw Місяць тому

      is English not your first language?

  • @BleachWizz
    @BleachWizz Місяць тому

    Oh no, this is going to make texts like I do!!!
    ok, drama aside, I do believe this will improve things a lot.
    I still see some caveats that would be left for luck, but huges amount of data might overcome that.
    I do believe we already have enough with GPT and a few previous ideas, still improving the language model itself is always a plus.

  • @Howoulduknow841
    @Howoulduknow841 Місяць тому

    This is something Anthropic has shared with their contextual retrieval.

  • @Isaacmellojr
    @Isaacmellojr Місяць тому

    Otima exemplificacao de como word2vec não é a solucao definitiva.

  • @АндрейАндреевич-з7т

    BM25. Frequency-weighted by sponsored-definition-tag vector search. Yeah google search do that too, you know. If you ever did seo optimization for your website or some kind of smm you know that it works

  • @pratikerande4808
    @pratikerande4808 Місяць тому

    super

  • @themax2go
    @themax2go Місяць тому +1

    ty for the insight to "pair" numerical rep (vector) w/ MB25... can the same be achieved w/ just using a knowledge graph? i'm experimenting w/ sci/phi triplex... what do you think, do you have any preliminary ideas, or have you already tested it and found using "entities_and_triples" not as effective / not effective at all? 6 mo ago you did a vid on knowledge graphs, i haven't watched it yet, i'll check it out...

  • @shizheliang2679
    @shizheliang2679 27 днів тому

    wait...I think I am in love...

  • @tempname-dr2bm
    @tempname-dr2bm 28 днів тому

    Poland mentioned

  • @bmm8213
    @bmm8213 Місяць тому

    Golden nugget

  • @NLPprompter
    @NLPprompter Місяць тому

    i love this bot...

  • @knucker3
    @knucker3 Місяць тому +4

    TURN YOUR VOLUME UP

  • @815TypeSirius
    @815TypeSirius Місяць тому

    But vs is enough to scam dummies and create a market bubble.

  • @ValidatingUsername
    @ValidatingUsername Місяць тому

    Try tokenizing engendered languages 😂

  • @rontheoracle
    @rontheoracle Місяць тому +12

    Excuse me, but your volume is just too low. Just saying.

    • @martin777xyz
      @martin777xyz Місяць тому +22

      Seems fine to me

    • @sladeTek
      @sladeTek Місяць тому +12

      No it’s not, your device is the issue

    • @rontheoracle
      @rontheoracle Місяць тому

      @@sladeTek It's just this video and a few others that play with very low volume. I try other videos in youtube, in general, they sound acceptably loud. Dunno why.

    • @rontheoracle
      @rontheoracle Місяць тому +1

      @@sladeTek Try watching the video in youtube with this title:
      "The Best RAG Technique Yet? Anthropic’s Contextual Retrieval Explained!"
      It is significantly much louder. Just my 2 cents.

    • @csmac3144a
      @csmac3144a Місяць тому +1

      Her audio is fine. Turn up your volume.

  • @Ruhgtfo
    @Ruhgtfo Місяць тому +1

    Contributed 3blue1brown