[LLM News] Moshi, RAG Best Practices, State of AI Report, Million Tiny Experts, GPT4All, RouteLLM

Elvis Saravia · Advanced ·🧠 Large Language Models ·1y ago

Another exciting episode of LLM News! Links mentioned in the video: 00:00 Moshi - https://x.com/kyutai_labs/status/1808883086173569222 01:54 Gen-3 Alpha - https://x.com/runwayml/status/1807822396415467686 02:19 RouteLLM - https://x.com/lmsysorg/status/1807812671238258931 04:10 Tiny Giant - https://x.com/SFResearch/status/1807811770267971984 05:23 Million Tiny Experts - https://x.com/omarsar0/status/1810389538340290724 06:31 1B Personas - https://x.com/omarsar0/status/1807827401122238628 07:53 Reasoning in LLMs - https://x.com/omarsar0/status/1810329294884741594 09:20 Best Practices in RAG - https://x.com/omarsar0/status/1808177231342018748 10:00 Self-Evaluation Defense - https://x.com/omarsar0/status/1809241930963853621 11:45 OpenAutoCoder - https://x.com/LingmingZhang/status/1808501612056629569 13:16 Satyrn - https://x.com/omarsar0/status/1810324765627867341 14:23 Pretzel - https://github.com/pretzelai/pretzelai/blob/main/README.md 15:10 GPT4All 3.0 - https://x.com/nomic_ai/status/1808162955806097767 17:15 Understanding Deep Learning - https://x.com/omarsar0/status/1808887392503279758 18:10 the State of AI Report - https://retool.com/blog/state-of-ai-h1-2024 #ai #machinelearning #science #engineering

Watch on YouTube ↗ (saves to browser)

Sign in to unlock AI tutor explanation · ⚡30

Playlist

Uploads from Elvis Saravia · Elvis Saravia · 0 of 60

← Previous Next →

101 ways to solve search (by Pratik Bhavsar)

101 ways to solve search (by Pratik Bhavsar)

TLDR Generation of Scientific Documents | ML Interview #1 with Isabel Cachola

TLDR Generation of Scientific Documents | ML Interview #1 with Isabel Cachola

Sentiment Analysis: Key Milestones, Challenges and New Directions

Sentiment Analysis: Key Milestones, Challenges and New Directions

Discriminative Adversarial Search for Abstractive Summarization (by Thomas Scialom)

Discriminative Adversarial Search for Abstractive Summarization (by Thomas Scialom)

Question Understanding: COVID-Q: 1,600+ Questions about COVID-19

Question Understanding: COVID-Q: 1,600+ Questions about COVID-19

Getting Started with NLP

Getting Started with NLP

Building tools and frameworks for large-scale social media mining (by Dr. Juan M. Banda)

Building tools and frameworks for large-scale social media mining (by Dr. Juan M. Banda)

TextAttack: A Framework for Data Augmentation and Adversarial Training in NLP

TextAttack: A Framework for Data Augmentation and Adversarial Training in NLP

Dive into Deep Learning (Study Group): Introduction to Deep Learning | Session 1

Dive into Deep Learning (Study Group): Introduction to Deep Learning | Session 1

Dive into Deep Learning (Study Group): Multilayer Perceptrons | Session 4

Dive into Deep Learning (Study Group): Multilayer Perceptrons | Session 4

How I read and annotate ML papers

How I read and annotate ML papers

Keep Learning ML (Session 1) | DSV, CompLex, Modern tools for emotions

Keep Learning ML (Session 1) | DSV, CompLex, Modern tools for emotions

Dive into Deep Learning (Study Group): Preliminaries | Session 2

Dive into Deep Learning (Study Group): Preliminaries | Session 2

Keep Learning ML #2 | Language-conditioned policy learning, Effective ML Testing, EagerPy

Keep Learning ML #2 | Language-conditioned policy learning, Effective ML Testing, EagerPy

Dive into Deep Learning (Study Group): Linear Neural Networks | Session 3

Dive into Deep Learning (Study Group): Linear Neural Networks | Session 3

Dive into Deep Learning (Study Group): Multilayer Perceptrons | Session 4

Dive into Deep Learning (Study Group): Multilayer Perceptrons | Session 4

Keep Learning ML #3 | Contrastively Trained Structured World Models

Keep Learning ML #3 | Contrastively Trained Structured World Models

Dive into Deep Learning (Study Group): Deep Learning Computation with PyTorch | Session 5

Dive into Deep Learning (Study Group): Deep Learning Computation with PyTorch | Session 5

Dive into Deep Learning (Study Group): Convolutional Neural Networks | Session 6

Dive into Deep Learning (Study Group): Convolutional Neural Networks | Session 6

Dive into Deep Learning (Study Group): Modern CNNs | Session 7

Dive into Deep Learning (Study Group): Modern CNNs | Session 7

101 ways to solve neural search with Jina

101 ways to solve neural search with Jina

(Hopefully-Reusable) Life Lessons for PhD Students in NLP

(Hopefully-Reusable) Life Lessons for PhD Students in NLP

How to save the world and forward your career in 5 easy steps | Women in NLP Talks

How to save the world and forward your career in 5 easy steps | Women in NLP Talks

Prompt Engineering Overview

Prompt Engineering Overview

Getting Started with the OpenAI Playground

Getting Started with the OpenAI Playground

LM-Guided Chain of Thought

LM-Guided Chain of Thought

Elements of a Prompt

Elements of a Prompt

Reasoning with Intermediate Revision and Search with LLMs #chatgpt #ai #llms #science #programming

Reasoning with Intermediate Revision and Search with LLMs #chatgpt #ai #llms #science #programming

General Tips for Designing Prompts

General Tips for Designing Prompts

Efficient Infinite Context Transformers #ai #machinelearning #research #llms #science

Efficient Infinite Context Transformers #ai #machinelearning #research #llms #science

Best Practices and Lessons Learned on Synthetic Data for Language Models #ai #machinelearning #genai

Best Practices and Lessons Learned on Synthetic Data for Language Models #ai #machinelearning #genai

Reducing Hallucinations in Structured Outputs via RAG #chatgpt #ai #llms #programming

Reducing Hallucinations in Structured Outputs via RAG #chatgpt #ai #llms #programming

Basic Prompt Examples for LLMs

Basic Prompt Examples for LLMs

LLM In Context Recall is Prompt Dependent #llms #ai #chatgpt #machinelearning

LLM In Context Recall is Prompt Dependent #llms #ai #chatgpt #machinelearning

Zero-shot Prompting Explained

Zero-shot Prompting Explained

RAG Faithfulness #llms #ai #gpt4

RAG Faithfulness #llms #ai #gpt4

Understanding LLM Settings

Understanding LLM Settings

Llama 3 is here! | First impressions and thoughts

Llama 3 is here! | First impressions and thoughts

Llama 3 is Here! #ai #llms #llama3

Llama 3 is Here! #ai #llms #llama3

Microsoft introduces Phi-3 | The most capable small language model?

Microsoft introduces Phi-3 | The most capable small language model?

Microsoft introduces Phi-3! #ai #llms #microsoft

Microsoft introduces Phi-3! #ai #llms #microsoft

Make Your LLM Fully Utilize the Context #ai #llms #machinelearning

Make Your LLM Fully Utilize the Context #ai #llms #machinelearning

When to Retrieve? #ai #llms #machinelearning

When to Retrieve? #ai #llms #machinelearning

Training an LLM to effectively use information retrieval

Training an LLM to effectively use information retrieval

State-of-the-art open-source LLM judges #ai #machinelearning #gpt4

State-of-the-art open-source LLM judges #ai #machinelearning #gpt4

Better and Faster LLMs via Multi-token Prediction

Better and Faster LLMs via Multi-token Prediction

AlphaMath Almost Zero #ai #science #machinelearning

AlphaMath Almost Zero #ai #science #machinelearning

SWE-Agent | An LLM-based Software Engineering Agent

SWE-Agent | An LLM-based Software Engineering Agent

[LLM NEWS] AlphaFold 3, xLSTM, OpenAI's Model Spec, DeepSeek-V2, OpenDevin CodeAct 1.0

[LLM NEWS] AlphaFold 3, xLSTM, OpenAI's Model Spec, DeepSeek-V2, OpenDevin CodeAct 1.0

LLM-powered tool for web scraping #ai #chatgpt #engineering

LLM-powered tool for web scraping #ai #chatgpt #engineering

Learn about LLMs in this NEW course #ai #chatgpt #engineering

Learn about LLMs in this NEW course #ai #chatgpt #engineering

[LLM NEWS] KANs, Gemma 10M Context, OpenAI Updates?, Automatic Prompt Engineering, Tokenizer Arena

[LLM NEWS] KANs, Gemma 10M Context, OpenAI Updates?, Automatic Prompt Engineering, Tokenizer Arena

[LLM News] GPT4-o, Project Astra, Veo, Copilot+ PCs, Gemini 1.5 Flash, Chameleon

[LLM News] GPT4-o, Project Astra, Veo, Copilot+ PCs, Gemini 1.5 Flash, Chameleon

Enhancing Answer Selection in LLMs #ai #machinelearning #engineering

Enhancing Answer Selection in LLMs #ai #machinelearning #engineering

On exploring LLMs #ai #promptengineering #chatgpt

On exploring LLMs #ai #promptengineering #chatgpt

Transformers Can Do Arithmetic with the Right Embeddings #ai #machinelearning #engineering

Transformers Can Do Arithmetic with the Right Embeddings #ai #machinelearning #engineering

[LLM News] xAI Series B, Codestral, LLM Guide, AutoGen Course, Symbolic Chain-of-Thought

[LLM News] xAI Series B, Codestral, LLM Guide, AutoGen Course, Symbolic Chain-of-Thought

PR-Agent #ai #gpt4 #software

PR-Agent #ai #gpt4 #software

Extracting features from Claude 3 Sonnet

Extracting features from Claude 3 Sonnet

Has prompt engineering been solved?

Has prompt engineering been solved?

Related AI Lessons

Anthropic Taps SpaceX's 220K-GPU Colossus 1 to Fix Claude Rate Limits

Anthropic partners with SpaceX to use their 220K-GPU Colossus 1 cluster to fix Claude API rate limits and improve performance

The Daimon Java SDK: Chat, Stream, and Query Memory from 3 Lines of Java

Simplify AI feature development in Java with the Daimon Java SDK, which provides a unified API for LLM inference and more

Unsloth + NVIDIA: 1.6x Faster LLM Fine-Tuning With 70% Less VRAM

Learn how Unsloth and NVIDIA's collaboration achieves 1.6x faster LLM fine-tuning with 70% less VRAM, and what this means for developers training on consumer GPUs

Dev.to · pickuma

“On Your Behalf”: The Four Words Rewriting the AI Industry

Discover how four words are transforming the AI industry with technologies like Remi, Orbit, and GPT 5.5

Chapters (15)

Moshi - https://x.com/kyutai_labs/status/1808883086173569222

1:54 Gen-3 Alpha - https://x.com/runwayml/status/1807822396415467686

2:19 RouteLLM - https://x.com/lmsysorg/status/1807812671238258931

4:10 Tiny Giant - https://x.com/SFResearch/status/1807811770267971984

5:23 Million Tiny Experts - https://x.com/omarsar0/status/1810389538340290724

6:31 1B Personas - https://x.com/omarsar0/status/1807827401122238628

7:53 Reasoning in LLMs - https://x.com/omarsar0/status/1810329294884741594

9:20 Best Practices in RAG - https://x.com/omarsar0/status/1808177231342018748

10:00 Self-Evaluation Defense - https://x.com/omarsar0/status/1809241930963853621

11:45 OpenAutoCoder - https://x.com/LingmingZhang/status/1808501612056629569

13:16 Satyrn - https://x.com/omarsar0/status/1810324765627867341

14:23 Pretzel - https://github.com/pretzelai/pretzelai/blob/main/README.md

15:10 GPT4All 3.0 - https://x.com/nomic_ai/status/1808162955806097767

17:15 Understanding Deep Learning - https://x.com/omarsar0/status/1808887392503279758

18:10 the State of AI Report - https://retool.com/blog/state-of-ai-h1-2024

5 Levels of AI Agents - From Simple LLM Calls to Multi-Agent Systems

Dave Ebbelaar (LLM Eng)