A Neuro-Symbolic Perspective on Large Language Models (LLMs)
Austria, Linz
Joined January 2023
- Tweets21
- Following32
- Followers79
- Likes35
SymbolicAI retweeted
Tokyo will become a Global R&D Powerhouse in AI.
Open AI is planning to open its first Asia based office in Tokyo.
..following the path of @SakanaAILabs?
bloomberg.com/news/articles/…
SymbolicAI retweeted
I've been working on QSTAR already back in 2014/2015.
Read the paper here:
sciencedirect.com/science/ar…
;)
SymbolicAI retweeted
Listen to our founders Marius & Leo talking about the symbolicAI framework - part 1. Interviewed by @johntanchongmin youtu.be/Ch9ygW62A34?t=165 #AIwednesday #deeplearning #machinelearning #LLM #generativeAI #Neurosymbolic #artificialintelligence #ai #datascience #innovation
Have a look at our latest release that brings the power of GPT with plugins from @ExtensityAI to your local shell pypi.org/manage/project/symb…
Or you could do the same and stay in your local dev environment by simply using the
@SymbolicAPI
framework and your favourite GPT engine or host your own backend. Here are some inspirational examples for shell tool interactions and Jupyter notebooks:
SymbolicAI retweeted
youtube.com/watch?v=R46Sskmm…
@DinuMariusC and @leojoinsthevoid share more about the ChatBot with tool use functionality in SymbolicAI. We're working towards an easy user interface to customize your own tools and memory storage/retrieval for your own ChatBot/agent!
SymbolicAI retweeted
Text-to-Audio Generation using Instruction-Tuned LLM and Latent Diffusion Model
latent diffusion model based approach (TANGO) outperforms the state-of-the-art AudioLDM on most metrics and stays comparable on the rest on AudioCaps test set, despite training the LDM on a 63 times smaller dataset and keeping the text encoder frozen. This improvement might also be attributed to the adoption of audio pressure level-based sound mixing for training set augmentation, whereas the prior methods take a random mix
abs: arxiv.org/abs/2304.13731
github: github.com/declare-lab/tango
SymbolicAI retweeted
Some said I am to optimistic when I showed this image three months ago and stated that SOON #GPT3-like #LLMs will be available at the core of the dev stack (@SymbolicAPI)
Well… you can now run GPT3-level #AI models on your laptop, phone, and Raspberry Pi arstechnica.com/information-…
SymbolicAI retweeted
Adding Conditional Control to Text-to-Image Diffusion Models
present a neural network structure, ControlNet, to control pretrained large diffusion models to support additional input conditions
paper: github.com/lllyasviel/Contro…
github: github.com/lllyasviel/Contro…
New release: v0.2.9
github.com/Xpitfire/symbolic…
Release notes:
Added Shell Apps (sym+)
Exposed SymbolicAI functions via shell commands
Improved documentation
Added Lambda Expression
and many more...
Read all details in our docs at github.com/Xpitfire/symbolic….
1
116
Here an overview of many text-to-🪄 representations 💪
generative ai 🗨️:
• text-to-gif (T2G)
• text-to-3D (T2D)
• text-to-text (T2T)
• text-to-NFT (T2N)
• text-to-code (T2C)
• text-to-image (T2I)
• text-to-audio (T2S)
• text-to-video (T2V)
• text-to-music (T2M)
• text-to-motion (T2Mo)
#artificialintelligence
102
Another nice example showcasing how text as a symbolic representation is essential for ML.
SceneScape: Text-Driven Consistent Scene Generation
abs: arxiv.org/abs/2302.01133
project page: scenescape.github.io/
text-driven perpetual view generation -- synthesizing long videos of arbitrary scenes solely from an input text describing the scene and camera poses
1
81
SymbolicAI retweeted
A tutorial from @OpenAI on how to build a question-answering system on your custom website
Preprocess:
1. Website -> text (Scraping)
2. Text -> chunks -> embed
Inference:
1. Question -> embed
2. Get topK similar chunks
3. Answer generation using 1 & 2
platform.openai.com/docs/tut…
Great thx @ai_austria for referencing us in their newsletter: aiaustria.substack.com/p/ai-…
New release: v0.2.7
github.com/Xpitfire/symbolic…
Release notes:
Added HuggingFace backend support for local LLM execution
Added server-cliend local model hosting
Added new test for local hosting
and many more...
1
67
SymbolicAI retweeted
Meet SymbolicAI: The Powerful Framework That Combines The Strengths Of Symbolic Artificial Intelligence (AI) And Large Language Models buff.ly/3wwBCM8 #DL #AI #ML #DeepLearning #ArtificialIntelligence #MachineLearning #ComputerVision #AutonomousVehicles