• Home
  • Technology
  • Gaming
  • Entertainment
  • World & Business
  • Science
  • Sports
  • AI
HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
  • HomeTechnologyGamingEntertainmentWorld & BusinessScienceSportsAI
    • Home
    • Technology
    • Gaming
    • Entertainment
    • World & Business
    • Science
    • Sports
    • AI
    AI
    Self-promo

    Cohere plans October 8 webinar on search and retrieval

    Cohere says its search modeling leaders will discuss RCP-nDCG@10, Embed 5 and the future of search and retrieval on X Live.

    CO
    NR
    2 Sources, 4h ago, first seen 4h ago

    TLDR

    Cohere plans an October 8 X Live webinar on RCP-nDCG@10, Embed 5 and search and retrieval. The company says RCP-nDCG@10 grades every result on its actual relevance, unlike traditional nDCG, which credits only results in a benchmark’s answer key. Cohere says human reviewers and MTEB back its results.

    Combined views

    722

    2 Sources, first seen 4h ago

    Combined views

    722

    2 Sources, first seen 4h ago

    4 likes
    4 likes
    1 comments
    1 saves

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Featured Source
    1 comments
    1 saves

    Sentiment

    Positive——Negative

    Summary

    Not enough discussion yet.

    No sentiment analysis available yet.

    Today's Rank

    —

    Not ranked yet

    Today's Rank

    —

    Not ranked yet

    2 Sources

    @cohereWant to hear more? Stay tuned for our webinar on Thursday, October 8 on X Live. Our search modelling leadership will be discussing RCP-nDCG@10, Embed 5, and the future of search and retrieval. https://cohere.com/register/rcp-ndcg-webinar4h
    @Nils_ReimersWe confirmed the approach in several human studies: Assume one embedding model has a higher traditional nDCG score than others => In only 53% of the cases humans agree that this model produces better search results, close to a coin flip. Compare this with RCP-nDCG: If a model gets a higher RCP-nDCG score, humans agree up to 97% that this model produces better search results.2h

    2 Sources

    @cohereWant to hear more? Stay tuned for our webinar on Thursday, October 8 on X Live. Our search modelling leadership will be discussing RCP-nDCG@10, Embed 5, and the future of search and retrieval. https://cohere.com/register/rcp-ndcg-webinar4h
    @Nils_ReimersWe confirmed the approach in several human studies: Assume one embedding model has a higher traditional nDCG score than others => In only 53% of the cases humans agree that this model produces better search results, close to a coin flip. Compare this with RCP-nDCG: If a model gets a higher RCP-nDCG score, humans agree up to 97% that this model produces better search results.2h