🤖 22 AI News EXPLAINED!!!

🚀 Timestamp 🚀: 00:00 Intro 00:21 MiniGPT-v2 01:33 Lemur 02:28 Octopus 04:14 AudioSep 06:03 Agent LTC 10:03 iTransformers for TS 11:01 LLM SFT Abilities 12:56 DeepSparse - LLM on CPU 17:05 LLM Truth Linear Structure 19:42 Show-1 - Text to Video 20:32 Replit AI Coder Model 21:30 Zephyr MIT License 22:28 Samantha and Dolphin 23:40 Arithmo Mistral 24:48 Attention Sinks 26:42 Agent Protocol 27:23 Meta Stable Signature 29:23 Adobe Content Credentials 31:25 Vesuvius Challenge Prize winners 34:19 Sam Altman Humane 34:45 OpenAI Revenue Skyrockets 35:25 Github Copilot $100M ARR 🔗 Links 🔗 apers & LLM Tools MiniGPT v2 https://minigpt-v2.github.io/ LEMUR: HARMONIZING NATURAL LANGUAGE AND CODE FOR LANGUAGE AGENTS https://huggingface.co/OpenLemur/lemur-70b-v1 https://arxiv.org/pdf/2310.06830.pdf Octopus: Embodied Vision-Language Programmer from Environmental Feedback https://choiszt.github.io/Octopus/ AudioSep - Separate Anything you Describe - https://github.com/Audio-AGI/AudioSep ADAPTING LLM AGENTS THROUGH COMMUNICATION https://arxiv.org/pdf/2310.01444v2.pdf ITRANSFORMER: INVERTED TRANSFORMERS ARE EFFECTIVE FOR TIME SERIES FORECASTING https://arxiv.org/pdf/2310.06625.pdf HOW ABILITIES IN LLMs ARE AFFECTED BY SUPERVISED FINE-TUNING DATA COMPOSITION https://arxiv.org/pdf/2310.05492.pdf Sparse Fine-tuned LLMs with DeepSparse - https://github.com/neuralmagic/deepsparse/tree/main/research/mpt https://huggingface.co/spaces/neuralmagic/sparse-mpt-7b-gsm8k THE GEOMETRY OF TRUTH: EMERGENT LINEAR STRUCTURE IN LARGE LANGUAGE MODEL REPRESENTATIONS OF TRUE/FALSE DATASETS https://arxiv.org/pdf/2310.06824.pdf Show-1: Marrying Pixel and Latent Diffusion Models for Text-to-Video Generation https://showlab.github.io/Show-1/ Replit’s new AI Model now available on Hugging Face https://blog.replit.com/replit-code-v1_5 Eric Hartford’s Dolphin and Samantha - https://huggingface.co/ehartford/dolphin-2.1-mistral-7b https://huggingface.co/ehartford/samantha-1.2-mistral-7b Zephry i