♊️ GemiNews 🗞️ (dev)

Demo 1: Embeddings + Recommendation Demo 2: Bella RAGa Demo 3: NewRetriever Demo 4: Assistant function calling

🗞️Daily Reading List – April 1, 2024 (#288)

🏷️ Daily Reading List

🗿Semantically Similar Articles (by :title_embedding)

Daily Reading List – April 1, 2024 (#288)

Article Image
2024-04-02 - Richard Seroter (from Richard Seroter Blog)

Today's links look at why it's risky to play it safe, how to use LLMs to improve factuality, and how you can build trust with others.

I had a wonderful Easter Sunday which set the tone for what will hopefully be an excellent week. Scan through a bunch of really solid items in today’s reading list. [paper] Long-form Factuality in Large Language Models. From DeepMind, this paper explains that larger language models are better for factuality, and LLM agents are better than humans at rating factual accuracy. [article] Why Playing It Safe Is the Riskiest Strategic Choice. The “timid transformation” may be worse than no transformation at all. You don’t have to be reckless when you decide to take bigger swings. [blog] GKE + Gemma + Ollama: The Power Trio for Flexible LLM Deployment. Open models run through open frameworks atop open platforms? Even when managed or proprietary services are “easier”, patterns like this will be justifiably popular for portability reasons. [blog] Using the platform engineering maturity model to understand the commitment required for an internal developer platform. Here’s a look at how you might tackle platform engineering. I’d recommend you don’t aim for super-maturity up front. Do a minimum viable platform and see what makes sense before over-investing and over-engineering. [blog] Avoid blundering: 80% of a winning strategy. Really good post from Jason that looks at the preventable mistakes that sink your business. [blog] Cloud CISO Perspectives: Get ready for Next ‘24: What you need to know. To make big conferences feel more controlled, it’s useful to carve up a domain of focus. If you care about security, there will be a lot to pay attention to at Google Cloud Next. Here are 15 sessions that are a good start. [blog] How To Know Who You Can Trust: 6 Secrets From Research. You won’t get far if people don’t trust you. But earning trust, and trusting others, isn’t easy. This post offers a few suggestions. [article] Visual Guide to Slices in Go. That is a LOT of coverage for slices (array abstractions), but I particularly liked the engaging visual treatment. It definitely made a potentially-dry topic come to life. [article] Redis vs. the trillion-dollar cabals. Matt is fired UP about the recent Redis fork and makes the argument that it wouldn’t be needed if cloud vendors pitched in more. More fork talk. [blog] Getting Started with Claude 3 on Google Cloud. Good look at trying out the latest Anthropic models. [blog] Generative AI to quantify uncertainty in weather forecasting. Is anyone going to tell me with a straight face that weather prediction is so good already that we don’t need AI? Didn’t think so. ## Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below: Subscribe

🏷️ Daily Reading List
[Blogs] 🌎 http://seroter.com/2024/04/01/daily-reading-list-april-1-2024-288/ [🧠] [v2] article_embedding_description: {:llm_project_id=>"Unavailable", :llm_dimensions=>nil, :article_size=>5274, :llm_embeddings_model_name=>"textembedding-gecko"}
[🧠] [v1/3] title_embedding_description: {:ricc_notes=>"[embed-v3] Fixed on 9oct24. Only seems incompatible at first glance with embed v1.", :llm_project_id=>"unavailable possibly not using Vertex", :llm_dimensions=>nil, :article_size=>5274, :poly_field=>"title", :llm_embeddings_model_name=>"textembedding-gecko"}
[🧠] [v1/3] summary_embedding_description: {:ricc_notes=>"[embed-v3] Fixed on 9oct24. Only seems incompatible at first glance with embed v1.", :llm_project_id=>"unavailable possibly not using Vertex", :llm_dimensions=>nil, :article_size=>5274, :poly_field=>"summary", :llm_embeddings_model_name=>"textembedding-gecko"}
[🧠] As per bug https://github.com/palladius/gemini-news-crawler/issues/4 we can state this article belongs to titile/summary version: v3 (very few articles updated on 9oct24)

🗿article.to_s 

------------------------------
Title: Daily Reading List – April 1, 2024 (#288)
Summary: Today's links look at why it's risky to play it safe, how to use LLMs to improve factuality, and how you can build trust with others.

[content]
I had a wonderful Easter Sunday which set the tone for what will hopefully be an excellent week.  Scan through a bunch of really solid items in today’s reading list.



[paper] Long-form Factuality in Large Language Models. From DeepMind, this paper explains that larger language models are better for factuality, and LLM agents are better than humans at rating factual accuracy.



[article] Why Playing It Safe Is the Riskiest Strategic Choice. The “timid transformation” may be worse than no transformation at all. You don’t have to be reckless when you decide to take bigger swings.



[blog] GKE + Gemma + Ollama: The Power Trio for Flexible LLM Deployment. Open models run through open frameworks atop open platforms? Even when managed or proprietary services are “easier”, patterns like this will be justifiably popular for portability reasons.



[blog] Using the platform engineering maturity model to understand the commitment required for an internal developer platform. Here’s a look at how you might tackle platform engineering. I’d recommend you don’t aim for super-maturity up front. Do a minimum viable platform and see what makes sense before over-investing and over-engineering.



[blog] Avoid blundering: 80% of a winning strategy. Really good post from Jason that looks at the preventable mistakes that sink your business.



[blog] Cloud CISO Perspectives: Get ready for Next ‘24: What you need to know. To make big conferences feel more controlled, it’s useful to carve up a domain of focus. If you care about security, there will be a lot to pay attention to at Google Cloud Next. Here are 15 sessions that are a good start.



[blog] How To Know Who You Can Trust: 6 Secrets From Research. You won’t get far if people don’t trust you. But earning trust, and trusting others, isn’t easy. This post offers a few suggestions.



[article] Visual Guide to Slices in Go. That is a LOT of coverage for slices (array abstractions), but I particularly liked the engaging visual treatment. It definitely made a potentially-dry topic come to life.



[article] Redis vs. the trillion-dollar cabals. Matt is fired UP about the recent Redis fork and makes the argument that it wouldn’t be needed if cloud vendors pitched in more. More fork talk.



[blog] Getting Started with Claude 3 on Google Cloud. Good look at trying out the latest Anthropic models.



[blog] Generative AI to quantify uncertainty in weather forecasting. Is anyone going to tell me with a straight face that weather prediction is so good already that we don’t need AI? Didn’t think so.



##



Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below:



		
			
				
					
						Subscribe
					
				
			
		
	
[/content]

Author: Richard Seroter
PublishedDate: 2024-04-02
Category: Blogs
NewsPaper: Richard Seroter Blog
Tags: Daily Reading List
{"id"=>4341,
"title"=>"Daily Reading List – April 1, 2024 (#288)",
"summary"=>"Today's links look at why it's risky to play it safe, how to use LLMs to improve factuality, and how you can build trust with others.",
"content"=>"\n

I had a wonderful Easter Sunday which set the tone for what will hopefully be an excellent week. Scan through a bunch of really solid items in today’s reading list.

\n\n\n\n

[paper] Long-form Factuality in Large Language Models. From DeepMind, this paper explains that larger language models are better for factuality, and LLM agents are better than humans at rating factual accuracy.

\n\n\n\n

[article] Why Playing It Safe Is the Riskiest Strategic Choice. The “timid transformation” may be worse than no transformation at all. You don’t have to be reckless when you decide to take bigger swings.

\n\n\n\n

[blog] GKE + Gemma + Ollama: The Power Trio for Flexible LLM Deployment. Open models run through open frameworks atop open platforms? Even when managed or proprietary services are “easier”, patterns like this will be justifiably popular for portability reasons.

\n\n\n\n

[blog] Using the platform engineering maturity model to understand the commitment required for an internal developer platform. Here’s a look at how you might tackle platform engineering. I’d recommend you don’t aim for super-maturity up front. Do a minimum viable platform and see what makes sense before over-investing and over-engineering.

\n\n\n\n

[blog] Avoid blundering: 80% of a winning strategy. Really good post from Jason that looks at the preventable mistakes that sink your business.

\n\n\n\n

[blog] Cloud CISO Perspectives: Get ready for Next ‘24: What you need to know. To make big conferences feel more controlled, it’s useful to carve up a domain of focus. If you care about security, there will be a lot to pay attention to at Google Cloud Next. Here are 15 sessions that are a good start.

\n\n\n\n

[blog] How To Know Who You Can Trust: 6 Secrets From Research. You won’t get far if people don’t trust you. But earning trust, and trusting others, isn’t easy. This post offers a few suggestions.

\n\n\n\n

[article] Visual Guide to Slices in Go. That is a LOT of coverage for slices (array abstractions), but I particularly liked the engaging visual treatment. It definitely made a potentially-dry topic come to life.

\n\n\n\n

[article] Redis vs. the trillion-dollar cabals. Matt is fired UP about the recent Redis fork and makes the argument that it wouldn’t be needed if cloud vendors pitched in more. More fork talk.

\n\n\n\n

[blog] Getting Started with Claude 3 on Google Cloud. Good look at trying out the latest Anthropic models.

\n\n\n\n

[blog] Generative AI to quantify uncertainty in weather forecasting. Is anyone going to tell me with a straight face that weather prediction is so good already that we don’t need AI? Didn’t think so.

\n\n\n\n

##

\n\n\n\n

Want to get this update sent to you every day? Subscribe to my RSS feed or subscribe via email below:

\n\n\n
\n\t\t
\n\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

\n\t\t\t\t\t\tSubscribe\n\t\t\t\t\t

\n\t\t\t\t
\n\t\t\t
\n\t\t
\n\t
",
"author"=>"Richard Seroter",
"link"=>"http://seroter.com/2024/04/01/daily-reading-list-april-1-2024-288/",
"published_date"=>Tue, 02 Apr 2024 00:26:56.000000000 UTC +00:00,
"image_url"=>"https://2.gravatar.com/avatar/2fab44e2f8decc2654cf4cb6b3bef813b072a16c7206880423c14d2e4be6f910?s=96&d=https%3A%2F%2F2.gravatar.com%2Favatar%2Fad516503a11cd5ca435acc9bb6523536%3Fs%3D96",
"feed_url"=>"http://seroter.com/2024/04/01/daily-reading-list-april-1-2024-288/",
"language"=>nil,
"active"=>true,
"ricc_source"=>"feedjira::v1",
"created_at"=>Tue, 02 Apr 2024 19:51:29.114489000 UTC +00:00,
"updated_at"=>Mon, 21 Oct 2024 18:05:53.198972000 UTC +00:00,
"newspaper"=>"Richard Seroter Blog",
"macro_region"=>"Blogs"}
Edit this article
Back to articles