Hikmat Samadov
@s3m3dov
Software Engineer with a passion for creating elegant and efficient solutions. Constantly learning and exploring new technologies to stay ahead. #tech #coding
You might like
I just posted a new entry: s3m3dov.substack.com/p/no-juniors-n… Is breaking into software development harder than ever? Why are junior roles disappearing, and what does it mean for the industry?
Every aspect of life is being stripped of color. Many have noticed this trend — but why exactly is it happening? Something deeper is going on… (thread) 🧵
5 Reasons Why No One Replies To Your Job Applications- How to Fix Them? We all make mistakes, it's true. But when looking for a new job, it's crucial to minimize those mistakes in your application process, as they may cost you the position.
8. Perplexity / @perplexity_ai ChatGPT on steroids 🤯 Get instant answers along with cited sources on any topic. 🔗 perplexity.ai
1. SciSpace Copilot Your AI research assistant. Do hours worth of reading and understanding in minutes. 🔗 scispace.com
BREAKING: Google opens its Generative AI Platform to all. Imagine unlocking the power of machine learning for everyone Businesses can now integrate Google's Vertex AI into their applications. Here's what you need to know:
Love this
"Most people die at 25 and aren't buried until they're 75" Why? 4 reasons and solutions:
1. Meta MusicGen The current explosion in AI has been over five vectors: coding, text, image, video, and voice. Meta just added a sixth: music. @DrJimFan called this week's release of MusicGen the "Llama moment for music AI." Listen for yourself:
10/ Imitating Reasoning Process of LLMs - develops a 13B model that learns to imitate the reasoning process of large foundational models like GPT-4; it leverages large-scale & diverse imitation data and surpasses Vicuna-13B.
Smaller LLMs Can Imitate Reasoning of Larger LLMs -13-billion param model learns from rich GPT-4 signals (explanations, step-by-step, complex instructions) guided by teaching of ChatGPT -Beats SoTA instruction-tuned LLM (Vicuna-13B) by 100% in reasoning arxiv.org/abs/2306.02707
7/ Fine-Grained RLHF - trains LMs with fine-grained human feedback; instead of using overall preference, more explicit feedback is provided at the segment level which helps to improve efficacy on long-form question answering and reduces toxicity.
F in RLHF is overall preference, which conveys limited info🙁 We introduce Fine-Grained RLHF🚀and train LMs with explicit feedback like "sentence 1 is not factual", "sentence 2 is toxic" More effective & enables LM customization arxiv.org/abs/2306.01693 finegrainedrlhf.github.io
4/ MusicGen - a simple and controllable model for music generation built on top of a single-stage transformer LM together with efficient token interleaving patterns; it can be conditioned on textual descriptions or melodic features.
We've just released MusicGen, and there is a @huggingface demo now, here is a thread about me playing with it just right now. huggingface.co/spaces/faceboo… A 🧵👇
5/ Augmenting LLMs with Databases - combines an LLM with a set of SQL databases, enabling a symbolic memory framework; completes tasks via LLM generating SQL instructions that manipulate the DB autonomously.
Augmenting LLMs with Databases - combines an LLM with a set of SQL databases, enabling a symbolic memory framework - completes tasks via LLM generating SQL instructions that manipulate the DB autonomously (supports selection, insertion, update, delete) arxiv.org/abs/2306.03901
3/ Sparse-Quantized Representation - a new compressed format and quantization technique that enables near-lossless compression of LLMs across model scales; “allows LLM inference at 4.75 bits with a 15% speedup”.
We present SpQR, which allows lossless LLM inference at 4.75 bits with a 15% speedup. You can run a 33B LLM on a single 24GB GPU fully lossless. SpQR works by isolating sensitive weights with higher precision and roughly doubles improvements from GPTQ: arxiv.org/abs/2306.03078🧵
2/ AlphaDev - a deep reinforcement learning agent which discovers faster sorting algorithms from scratch; the algorithms outperform previously known human benchmarks and have been integrated into the LLVM C++ library.
Just In! DeepMind presents AlphaDev, a deep reinforcement learning agent which discovers faster sorting algorithms from scratch. The algorithms outperform previously known human benchmarks and have been integrated into the LLVM C++ library. nature.com/articles/s4158…
Why Companies NEED People Back In The Office? youtu.be/jrsRvozsUQ8
youtube.com
YouTube
Why Companies NEED People Back In The Office
Google Strikes Back (The AI Wars) youtu.be/Qa4K7XsRO0g
youtube.com
YouTube
Google Strikes Back (The AI Wars)
"We don't support virtual environments inside WSL. There is a related ticket in our issue tracker about that" youtrack.jetbrains.com/issue/PY-32853 youtrack.jetbrains.com/issue/PY-52688….
United States Trends
- 1. Araujo 177K posts
- 2. Chelsea 612K posts
- 3. Barca 246K posts
- 4. Estevao 244K posts
- 5. Barcelona 440K posts
- 6. Ferran 75.8K posts
- 7. Yamal 188K posts
- 8. Oklahoma State 4,449 posts
- 9. Eric Morris 2,949 posts
- 10. Cucurella 95.4K posts
- 11. Flick 46.8K posts
- 12. Godzilla 24.9K posts
- 13. Rashford 24.1K posts
- 14. Kounde 48.1K posts
- 15. Pedri 52K posts
- 16. Enzo 66.3K posts
- 17. Skippy 4,369 posts
- 18. #CHEBAR 16.6K posts
- 19. Raising Arizona 1,720 posts
- 20. Happy Thanksgiving 24.8K posts
You might like
Something went wrong.
Something went wrong.