TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Rerank 3: A new foundation model for efficient enterprise search and retrieval

45 pointsby bguberfainabout 1 year ago

3 comments

dvtabout 1 year ago
Being as charitable as possible here, and Rerank 3 might be the bee&#x27;s knees, but the examples are absolutely <i>awful</i>. Do you really need to use embeddings + a large language model to search for &quot;action&quot; and &quot;Christian Bale&quot; in two columns[1]?<p>Your interface can literally just be two dropdowns. I&#x27;d like to see things like &quot;the actor that played the Joker in that movie about Bob Dylan&quot; if you&#x27;re really trying to flex your semantic search muscles.<p>[1] <a href="https:&#x2F;&#x2F;colab.research.google.com&#x2F;drive&#x2F;1sKEZY_7G9icbsVxkeEIA_qUthEfPrK3G?usp=sharing&amp;ref=txt.cohere.com" rel="nofollow">https:&#x2F;&#x2F;colab.research.google.com&#x2F;drive&#x2F;1sKEZY_7G9icbsVxkeEI...</a>
esafakabout 1 year ago
Someone correct me if I&#x27;m mistaken, but Cohere appears to be using BM25 and semantic search (Embed Multilingual) individually as baselines in order to look better. A more suitable baseline would be the Reciprocal rank fusion (RRF) of BM25 and semantic search. And those latencies seem high; seconds to rerank?
bigcat12345678about 1 year ago
El5 me what is rerank model? Why 4k context window size is considered large?
评论 #40005380 未加载