Kauta Marvin

Developer

[Claude @claudeai]

Introducing Claude Fable 5: a Mythos-class model that we’ve made safe for general use.

Its capabilities exceed those of any model we’ve ever made generally available.

[Claude @claudeai]

This is Claude Sonnet 4.6: our most capable Sonnet model yet.

It’s a full upgrade across coding, computer use, long-context reasoning, agent planning, knowledge work, and design.

It also features a 1M token context window in beta.

[Howwebiz @HowweEnt]

So sad 😢 @saadShots 😭😭

[Kauta Marvin @KautaMavin]

@Onlineboss_ genwealth

[The Crypto Dog 📈 @TheCryptoDog]

Study ETHBTC
Study Dec 2016 Jan 2017
Study #altseason

ETH $6 to $1400 that year
Now extrapolate to your altcoins

[Kauta Marvin @KautaMavin]

@CryptoNobler 5W15mGTpmVM8sGGftx2iHZ1c6KxDge9H82skZLLkwqEy

[Kauta Marvin @KautaMavin]

@_TJRTrades 2Fb9u1hWq8rcFXZSWy285jWaJ4LmtXEQFUZQFhAG2qY6

[Kauta Marvin @KautaMavin]

@little_mugisu @KhanzJoel @UCC_Official @UgCERT https://t.co/2fMnxUK7g7

[Joel. @KhanzJoel]

The ongoing @UCC_Official @UgCERT Cyber drill has seen over 195 cyber security enthusiasts both beginners and professionals have hands on hacking competition as they solve numerous simulated cyber security challenges. As it stands now the top ten are shown below.

[CAF @CAF_Online]

Two worldies in two minutes, Uganda caught South Africa off guard last night. 🚀🇺🇬

#AFCONQ2025

[NBS Television @nbstv]

BREAKING: NUP President Robert Kyagulanyi (Bobi Wine) has reportedly been shot in the leg.

More updates to follow.

#NBSUpdates

[LangChain @LangChain]

💪RAG Me Up

RAG Me Up is a generic framework (server + UIs) that enables you do to RAG on your own dataset easily. Its essence is a small and lightweight server and a couple of ways to run UIs to communicate with the server (or write your own).

https://t.co/zWo4RGHdNp

[Kauta Marvin @KautaMavin]

marco,a project I'm developing in my nights and weekends. its all about event discovery on WhatsApp . With Marco, you can effortlessly discover, manage, and engage with events. Stay tuned for more updates
@_nightsweekends

[Kauta Marvin @KautaMavin]

@AndrewBolis Guide

[Kauta Marvin @KautaMavin]

@matchaman11 Am Interested but not frok
India

[Uganda Christian University (Official) @UCUniversity]

UCU emerges victorious at the 1st National Council for Higher Education Regional Exhibition hosted at UCU Mbale University College.

[UCU Computing, Engineering, Data Science, Robotics @ucu_ComputEng]

New multidisciplinary short courses open for April, May, and June intake

Apply now using the link below.

https://t.co/GRqezTdaVX

April Intake deadline: 29th March 2024

For details, WhatsApp us on 0708114300 or Call us on 0312350856

[LangChain @LangChain]

1/ 🌲Language Agent Tree Search (LATS)

LATS is a unified algorithm that uses self-reflection (and additional external feedback) to self-optimize and efficiently adapt to new tasks and environments. It performs better than similar algorithms like Tree of Thoughts, ReAct, and Reflexion.

LATS unifies a few important trends in AI:

1. Search (algorithmic)
2. Planning & "Reasoning"
3. Executing / Acting

It does this without any updates to the model (unlike complicated RL).

We created a simple implementation of LATS in LangGraph to demystify things, so you can appropriately apply it in your applications.

Check out the technical walkthrough and other results in the thread:

Paper by @andyz245, @aiatillinois,@HaohanWang, et. al.

[LangChain @LangChain]

Self RAG

Self-reflection is one of the most interesting ideas in RAG, giving an LLM the ability to self-correct problems in retrieval and / or generation.

But, implementing this reliably can be tricky. We've found that graphs are a great way to reliably express logical flows, such as self-reflective RAG.

Here's a cookbook that shows how to implement ideas from the self-RAG paper by (@AkariAsai et al) using LangGraph.

It first uses a retrieval grader to (1) assess the quality of retrieved documents relative to the query. It then grades generations both for (2) hallucinations and for (3) usefulness in answering the question.

Cookbook:
https://t.co/3ThXzscI7g

Paper:
https://t.co/2oxTQOtViK

[LangChain @LangChain]

RAG From Scratch: Query Translation (Multi-Query)

Our RAG From Scratch video series walks through impt RAG concepts in short / focused videos w/ code.

Over the next few days we'll release videos focused on Query Translation, starting with Multi-Query.

🔧 Problem: User queries are a challenge in RAG. If a user provides an ambiguous query, they can retrieve ambiguous documents w/ distance-based similarity search.

💡 Idea: Re-write the user question from multiple perspectives, retrieve documents for each re-written question, return the unique documents for all queries.

📽️ Video:
https://t.co/7aqRET7NKw

💻 Code:
https://t.co/DGogC0sEZc

🧠 Related works / ideas:
1/ Ma et al w/ RRR
https://t.co/LWzW7e5DVt
2/ @Raudaschl w/ RAG-Fusion
https://t.co/sQvYmcnHXb
3/ @RickLamers
https://t.co/VcNt24Q8xy