PaperMatch

Languages and libraries

semantic paper search

SHOW HN2024

324points
104comments
#175of 3,515 in 2024
Top 5%of 2024 by points

How HN took it

MixedNo clear lean

16 positive, 18 neutral or asking, 2 mixed, 9 critical, out of 45 comments about the project. Thread score +0.13.

Idea: 9 for, 1 againstExecution: 1 for, 5 againstPerformance: 0 for, 3 against

The thread, comment by comment

Each row is a top-level comment and its replies, in HN's order. Select a cell to open that comment on HN.

positive neutral or mixed critical about another commenter or something else the maker

What HN said

What people liked

  • Semantic results are useful for discovering related papers

    From 2 comments
    • This is great! I just tried some queries and the results were pretty decent, in terms of semantics. But, just thinking of it as a user, if this were to be part of my daily…
      curious_cat_163 on HN
    • All benefits depend on the ability of the embedding model. Semantic embeddings understand nuances, so they can match abstracts that align conceptually even if no exact keywords…
      Quizzical4230 on HN
  • Useful companion to bioRxiv and medRxiv versions

    From 1 comment
    • This looks great! I have used the biorXiv version of papermatch and it gives pretty good results!
      madbutcode on HN

Objections

  • Searches for specific terms like leaky relu return irrelevant results

    From 2 comments
    • Feedback: first thing I tried is searching for "leaky relu" and I got a bunch of results related to fluids, which is... not very relevant. (: Compare that to scholar which…
      kouteiheika on HN
    • This seems like a cool idea, thanks for creating it! Some feedback: I tried searching for "wave function collapse algorithm", "gumin wave function collapse", "wfc" and "model…
      zzyzek on HN
  • Author search does not work

    From 1 comment
    • I tried a simple search by author and it didn’t work. All the fancy stuff is great, but I’d expect the basics still work, in the end it’s a search engine for papers.
      serial_dev on HN
  • Cloudflare challenge screen is a barrier

    From 1 comment
    • > The Cloudflare challenge screen at the beginning is a dealbreaker. I understand your concern, however, I do not have the know-how to properly combat bots that keep spamming the…
      Quizzical4230 on HN
  • Binary embeddings reduce recall beyond top 10 results

    From 2 comments
    • That looks great for speed, but what about recall?
      binarymax on HN
    • That's has a major downgrade. For binary embeddings, the top 10 results are same as fp32, albeit shuffled. However after the 10th result, I think quality degrades quite a bit. I…
      Quizzical4230 on HN

Requests

  • Date or recency filter

    From 2 comments
    • I couuld and really use this, but it didn't work for me. And HAS to have a date filter. That is a must maybe with some time based pre-option defaults like HackerNews. Good luck,…
      ProofHouse on HN
    • 1. why mixbread's model? 2. how much efficiency gain did you see binarising embeddings/using hamming distance? 3. why milvus over other vector stores? 4. did you automate the…
      swyx on HN
  • Keyword and quoted-phrase search

    From 1 comment
    • Exactly. Full text paradigm has it's own pros and I believe we need those tools in the new vector search to take full advantage. I am planning to add keywords feature where if a…
      Quizzical4230 on HN
  • Similar papers link on each result

    From 1 comment
    • Very cool! Add a "similar papers" link to each paper, that will make this the obvious way to discover topics by clicking along the similar papers.
      Maro on HN
  • Fix LaTeX rendering in abstracts

    From 1 comment
    • interesting project; I’m not really sure how useful it is for field-specific stuff—I'm searching for “image reduction astronomy”, and it shows all sorts of related but not…
      venice_benice on HN

Questions

  • Why binarised embeddings and Hamming distance

    From 2 comments
    • Nice work. Any other technical comments, why did you use those embeddings, did you binarzue them, did you use any dpecial prompts?
      antman on HN
    • Did you notice a difference in performance after binarization? Do you have a way to measure performance?
      andai on HN
  • How search quality is evaluated

    From 1 comment
    • It sounds nice. How do you evaluate the performance of your way against usual embedding?
      higty on HN

Alternatives named

  • Semantic Scholar

    From 1 comment
    • Is this similar to https://www.semanticscholar.org (from Allen Institute for AI) ?
      shigeru94 on HN
  • Google Scholar

    From 1 comment
    • Feedback: first thing I tried is searching for "leaky relu" and I got a bunch of results related to fluids, which is... not very relevant. (: Compare that to scholar which…
      kouteiheika on HN
  • Research Rabbit

    From 1 comment
    • This looks great, thanks for building this. Something on similar lines which many may link, Research Rabbit - https://www.researchrabbit.ai/
      zerop on HN
  • Elicit, Undermind, scite

    From 1 comment
    • I enjoy seeing projects like this! If you expand beyond arxiv, keep in mind since coverage matters for lit reviews, unfortunately the big publishers (Elsevier and Springer) are…
      shishy on HN

Category, keywords, summaries and labels on this page were made by AI (Thread Reader v2) and not checked by hand. Each point links to the comments it came from.

Add a badge to your READMEShown on HackerShow
[![Shown on HackerShow](https://hackershow.org/badge.svg)](https://hackershow.org/show/42507116)