singularit

Towards the AI ​​Singularity 🌠
1st Singularit Dev Awards 🏆 Dec/24

[NeurIPS Conference @NeurIPSConf]

Please read our statement on the remarks made by Dr. Rosalind Picard at her NeurIPS 2024 invited talk and our commitment to respect, inclusivity, and upholding our values:
https://t.co/AsAtriVq5V

[Utopic e/λ @UtopicDev]

🤷‍♂️

[Nous Research @NousResearch]

Nous Research announces the pre-training of a 15B parameter language model over the internet, using Nous DisTrO and heterogeneous hardware contributed by our partners at @Oracle, @LambdaAPI, @NorthernDataGrp, @CrusoeCloud, and the Andromeda Cluster.

This run presents a loss curve and convergence rate that meets or exceeds centralized training.

Our paper and code on DeMo, the foundational research that led to Nous DisTrO, is now available (linked below).

[Vaibhav (VB) Srivastav @reach_vb]

Introducing Indic-Parler TTS - Trained on 10K hours of data, 938M params, supports 20 Indic languages, emotional synthesis, apache 2.0 licensed! 🔥

A collaboration w/ @ai4bharat & @huggingface - w/ fully customisable speech and voice personas!

Try it out directly below or use the model weights as you want!

🇮🇳/acc

[Arash Vahdat @ArashVahdat]

📢 Warped Diffusion

Our new #neurips2024 presents a simple approach to turn image2image models into video2video models.

https://t.co/NQIQSbiO3y
https://t.co/X6IjpewloR

with: @giannis_daras, @wn8_nie, @karsten_kreis , @AlexGDimakis , @MardaniMorteza, Nik Kovachki

[AK @_akhaliq]

Husky

A Unified, Open-Source Language Agent for Multi-Step Reasoning

Language agents perform complex tasks by using tools to execute each step precisely. However, most existing agents are based on proprietary models or designed to target specific tasks, such as

[SkalskiP @skalskip92]

supervision - a computer vision library I created - just crossed 15,000 stars on GitHub!

BBBRRRRRR!

link: https://t.co/xXMRaS3Guk

[Gradio @Gradio]

🤯 MimicBrush might be the coolest AI app on 🤗 Spaces right now!

In-the-wild results are really something!🚀

[João Moura @joaomdmoura]

🔥🔥 crewAI v0.30.10 is out🔥🔥

🧙 Set a manager
📄 Set system, prompt and response templates
👨‍💻 Improved json/pydantic output
🔎 Better tool recognition
🧰 Better tool usage
📄Bring your prompts
🪚 New tools @browserbase and @ExaAILabs
And more! Docs will follow

RT please?

[Div @divyanshu_van]

We need to talk about Agents! Inspired by @llama_index's recent tutorials on Agents (@ravithejads) and @jerryjliu0's talk on Advanced Context-Augmented LLM
Applications, this post looks at the Planning component of an agentic workflow. Happy Reading! https://t.co/kGD7nFrw0z

[Python Trending 🇺🇦 @pythontrending]

efficient-kan - An efficient pure-PyTorch implementation of Kolmogorov-Arnold Network (KAN). https://t.co/pPrqqDFpij

[LangChain @LangChain]

⚡️Multi-Agent RAG Online Workshop⚡️

If there’s anything better than agentic RAG, it’s multi-agent RAG!

In this event, we’ll explore the big idea behind “multi-agent” applications. These types of workflows combine multiple independent agents, which can be structured to work together to solve complex problems.

The patterns of using multi-agent frameworks can take many forms including planning, reflection, and tool use. LangGraph, our library for building stateful, multi-actor applications with LLMs, can help you orchestrate your system.

Sign up here ➡️ https://t.co/QBaPPnGOhR

[LangChain @LangChain]

📄Prompt Engineering with LangChain

This LinkedIn course by @DataScienceHarp is a great resource for getting started

It's over 5 hours long, and Harpreet has been a fantastic member of the community since the early days. Big thanks for all the effort!

https://t.co/59873Egkpm

[merve @mervenoyann]

Parameter-free LLaVA for video captioning works like magic! 🤩
Let's take a look 🧶
(also find repositories below ⬇️ )

[Gradio @Gradio]

🤯ByteDance's official StoryDiffusion demo is already out on @huggingface Spaces!

🙌The comics-related codes are public now. Kudos to @zhoudaquan21 et al. for the great work. Links and more 👇

[LlamaIndex 🦙 @llama_index]

Check out this neat video from @jasonzhou1993 which gives a nice overview of the components needed for agentic RAG:

✅ LlamaParse + Firecrawl for advanced document processing
✅ An overview of agentic RAG techniques

https://t.co/xBFyFNZuaM

If you’re interested in building agentic RAG check out @jerryjliu0’s talk! https://t.co/x0z8KfUx6o

[Jerry Liu @jerryjliu0]

This is a fantastic papercard by @_nerdai_ on CRITIC (@zebgou et al.) - using external tools to critique and improve the output of LLM agents.

In comparison to other reflection approaches, CRITIC relies on external tools (e.g. perspective API for toxicity reduction, code interpreter to validate/run code), to generate feedback about a current agent’s trajectory and output. This offers an element of getting external feedback; in many cases this gives much better results than only using LLMs to critique itself.

Best of all, we have full YouTube walkthroughs, notebooks, and @llama_index abstractions so that you can either apply or build a CRITIC approach to agent/RAG pipeline that you’re building!

YouTube walkthrough: https://t.co/yh2o3qQuVb
Notebook: https://t.co/6R27uMFp77
CRITIC agent on LlamaHub 🦙🏡: https://t.co/r4gGBNaOJz
Source paper: https://t.co/zxfFwUDG6c

[Alex Reibman 🖇️ @AlexReibman]

New Tutorial: AI agents to automatically create feedback reports from YouTube comments by @tonykipkemboi

Built with @crewAIInc + @AgentOpsAI + @GroqInc + @ollama

Full video👇:

[Alvaro Cintas @dr_cintas]

This new model on Replicate allows you to virtually try on clothes.

All you need is an image of the person and the clothing to ‘try on’.

Here is how you can do it: