Top all-time · #2789
Cohere
Company blog · txt.cohere.comaiprogramming
Cohere is the company blog at txt.cohere.com, writing about ai and programming. Since 2023, 27 different Hacker News readers have linked it 39 times, most often “Text Embeddings Visually Explained (2022)”. No feed was found, so read it on the site itself.
Explore our collection of insightful blog posts covering a diverse range of generative AI topics. Our articles offer in-depth analyses, expert opinions, and practical advice to inform and inspire.
Most-linked posts?
- Text Embeddings Visually Explained (2022)4 people · 5 mentions · first linked 2023-06-25
- Embedding Archives: Millions of Wikipedia Article Embeddings in Many Languages3 people · 3 mentions · first linked 2023-04-20
- Embed v3: Cohere's latest and most advanced text embedding model2 people · 2 mentions · first linked 2023-11-02
- Command-R: RAG at production scale2 people · 2 mentions · first linked 2024-03-11
- Say Goodbye to Irrelevant Search Results: Cohere Rerank Is Here2 people · 2 mentions · first linked 2023-05-01
- Cohere Compass Private Beta: A New Multi-Aspect Embedding Model2 people · 2 mentions · first linked 2024-04-15
What Hacker News says?
This technique had a very recent resurgence via https://txt.cohere.com/int8-binary-embeddings/ . Hugging face also covered the technique here https://huggingface.co/blog/embedding-quantization . It seems like a very good tradeoff compared to the shorter embeddings which require fine tuning via the matryoshka technique. On the other hand, Nils Reimers suggests that trivial quantization of the full precision embedding…
Comment linking /int8-binary-embeddings/
on “Binary vector search is better than FP32 vectors”Submitted Command R+: A Scalable LLM Built for Business as a story · 59 points
HN discussionI'm the author of https://jalammar.github.io/illustrated-transformer/ and have spent years since introducing people to Transformers and thinking of how best to communicate those concepts. I've found that different people need different kinds of introductions, and the thread here includes some often cited resources including: https://peterbloem.nl/blog/transformers https://e2eml.school/transformers.html I would also…
Comment linking /what-are-transformer-models/
on “Ask HN: Can someone ELI5 Transformers and the “Attention is all we need” paper”Submitted Rerank 3: A new foundation model for efficient enterprise search and retrieval as a story · 45 points
HN discussionProject Aya is one such attempt at a multi-lingual model (targeting 101 languages): - https://txt.cohere.com/aya-multilingual/ - https://aya.for.ai I'm a contributor to the project and all data and model will be open-sourced. We're looking for contributors in many languages!
Comment linking /aya-multilingual/
on “Finland's ChatGPT equivalent begins to think in Estonian as well”Submitted Command-R: RAG at production scale as a story · 37 points
HN discussionhttps://txt.cohere.com/command-r/
Comment linking /command-r/
on “Command-R, a new RAG-optimized LLM aimed at large-scale production workloads”Submitted LLM University – Your Go-To Learning Resource for NLP as a story · 8 points
HN discussionYes -- this is a good article on it: https://txt.cohere.com/sentence-word-embeddings As a blackbox, it is a generalization of word2vec to sequence2vec. For example, simply summing or averaging word vectors in a sentence can give you a fast & cheap sentence embedding. But natural language sentences have more structure than natural language words. Ex: it matters precisely where "not" goes in a sentence. So a lot of im…
Comment linking /sentence-word-embeddings
on “Emerging Architectures for LLM Applications”Submitted Say Goodbye to Irrelevant Search Results: Cohere Rerank Is Here as a story · 4 points
HN discussionThere's this by Cohere - https://txt.cohere.com/embedding-archives-wikipedia/
Comment linking /embedding-archives-wikipedia/
on “Having a 20GB file that lets you ask an offline computer any question is amazing”Submitted Embedding Archives: Millions of Wikipedia Article Embeddings in Many Languages as a story · 4 points
HN discussionThere's a lot you can do with the vectors themselves without needing to embed any more text (e.g., clustering, exploration, visualization after dimensionality reduction...etc). Here's a previous embeddings exploration of top HN posts: https://txt.cohere.com/combing-for-insight-in-10-000-hacker-... A lot of that code can be used here as well. If you want to query for a search term, you can use a trial API key which i…
Comment linking /combing-for-insight-in-10-000-hacker-news-posts-with-text-clustering/
on “Embedding Archives: Millions of Wikipedia Article Embeddings in Many Languages”Submitted Building a RAG chatbot with query rewriting and citations as a story · 3 points
HN discussionProject Aya is a multi-institution collaborative project to accelerate multi-lingual AI and make sure diverse languages are properly represented in this wave of large language models. The first release is targeting 101 languages and we're 7 weeks away from starting to package it. I'm one of the contributor to the project and all data and model will be open-sourced. We're looking for more contributors across many lan…
Comment linking /aya-multilingual/
on “Aya: An open science project to build open multilingual models and datasets”Submitted Cohere Compass Private Beta: A New Multi-Aspect Embedding Model as a story · 2 points
HN discussionSince publishing this I've found a few additional resources that are really useful for understanding embeddings at a lower level (my article is deliberately very high level and focuses mainly on their applications). Cohere's Text Embeddings Visually Explained: https://txt.cohere.com/text-embeddings/ The Tensorflow Embedding Projector tool: https://projector.tensorflow.org/ What are embeddings? by Vicki Boykis is wor…
Comment linking /text-embeddings/
on “Embeddings: What they are and why they matter”Submitted Cohere for AI Launches Aya, an LLM Covering More Than 100 Languages as a story · 2 points
HN discussionProject Aya is a multi-institution collaborative project to accelerate multi-lingual AI and make sure diverse languages are properly represented in datasets and this wave of large language models. The first release is targeting 101 languages. I'm one of the contributor to the project and all data and model will be open-sourced. We're looking for more contributors across many languages! It’s quite easy to contribute.…
Comment linking /aya-multilingual/
on “Aya: An open science project to build open multilingual models and datasets”Submitted Text Embeddings Visually Explained (2022) as a story · 2 points
HN discussionProject Aya is a multi-institution collaborative project to accelerate multi-lingual AI and make sure diverse languages are properly represented in this wave of large language models. The first release is targeting 101 languages. I'm one of the contributor to the project and all data and model will be open-sourced. We're looking for more contributors across many languages! It’s quite easy to contribute. No knowledge…
Comment linking /aya-multilingual/
on “Aya: An open science project to build open multilingual models and datasets”Submitted Embed v3: Cohere's latest and most advanced text embedding model as a story · 2 points
HN discussionGiven the title, I was expecting a conceptual deep-dive into why and how embeddings are magical. The bulk of the content is more along the lines of a how-to guide for getting started with embeddings. The best post I've seen so far that really does dive into the magic of embeddings is this: https://txt.cohere.com/text-embeddings/
Comment linking /text-embeddings/
on “Stack: The Magic of Embeddings”Submitted We Built an AI-Powered Magic the Gathering Card Generator as a story · 2 points
HN discussionCheck Cohere's embeddings of Wikipedia: https://txt.cohere.com/embedding-archives-wikipedia/
Comment linking /embedding-archives-wikipedia/
on “a thread”Submitted Aya: An open science project to build open multilingual models and datasets as a story · 2 points
HN discussionSubmitted OpenAI Competitor Cohere Announces $270M Series C as a story · 2 points
HN discussionSubmitted Say Goodbye to Irrelevant Search Results: Cohere Rerank Is Here as a story · 2 points
HN discussionSubmitted Cohere Compass Private Beta: A New Multi-Aspect Embedding Model as a story · 1 points
HN discussionSubmitted Text Embeddings Visually Explained (2022) as a story · 1 points
HN discussion
