AI, Quant 투자 공부

https://medium.com/@ignacio.de.gregorio.noblejas/is-tiktok-planning-an-infinite-memory-chatgpt-c195b1a6eced Is TikTok Planning An Infinite Memory ChatGPT? ByteDance Research Hints At It medium.com Unleashing Infinite-Length Input Capacity for Large-scale Language Models with Self-Controlled Memory System https://arxiv.org/pdf/2304.13343.pdf 바이트댄스는 SCM (Self-Controlled Memory)를 발표하고 챗봇을 무한 입력이 가능..

→https://medium.com/@simon_attard/leveraging-large-language-models-in-your-software-applications-9ea520fb2f34 Leveraging Large Language Models in your Software Applications How can you leverage the capabilities of Large Language Models (LLMs) within your software applications? medium.com Overview LLM 위에 단순히 얇은 응용 프로그램을 구축하는 것은 다음과 같은 문제가 있음 사용자에 대한 응답은 예측할 수 없으며 환각을 포함 응답은 애플리케이션의 데이터 및 사용 사례에 근..

1. Vector databases source: https://medium.com/aimonks/vector-databases-7d46054e933 2. Leveraging Large Language Models in your Software Applications source: https://medium.com/@simon_attard/leveraging-large-language-models-in-your-software-applications-9ea520fb2f34 3. GPT in 60 Lines of NumPy source: https://jaykmody.com/blog/gpt-from-scratch/#gpt-architecture 4. hatGPT의 전두엽(장기기억 저장소)으로 각광받고 있는..

https://a16z.com/2023/06/20/emerging-architectures-for-llm-applications/ Emerging Architectures for LLM Applications | Andreessen Horowitz A reference architecture for the LLM app stack. It shows the most common systems, tools, and design patterns used by AI startups and tech companies. a16z.com 디자인 패턴: In-Context Learning In-Context Learning의 핵심 아이디어 Fine-tuning 없이 LLMs를 사용 대신 개인 컨텍스트 데이터에 대한 영..

1. Generative AI - Document Retrieval and Question Answering with LLMs source: https://medium.com/google-cloud/generative-ai-document-retrieval-and-question-answering-with-llms-2b0fb80ae76d Fine-tuning vs Indexing New Documents Fine-tuning은 몇 시간이 걸리는 반면, Indexing은 실시간으로 이용할 수 있다. Context Size Limiation 대부분의 LLM은 4K 토큰만을 허용하므로 많은 양의 데이터를 제공할 수 없음. Indexing 접근 방식을 사용하면 관련 문서와 유사한 문서를 검색. LLM은 무제한 ..

1. AI Canon source: https://a16z.com/2023/05/25/ai-canon/ 2. State of GPT source: https://build.microsoft.com/en-US/sessions/db3f4859-cd30-4445-a0cd-553c3304f8e2 3. VOYAGER: An Open-Ended Embodied Agent with Large Language Models source: https://arxiv.org/pdf/2305.16291.pdf 4. LLaMA-Adapter: Efficient Fine-tuning of Language Models with Zero-init Attention source: https://arxiv.org/pdf/2303.1619..

1. Run Very Large Language Models on Your Computer source: https://medium.com/p/390dd33838bb 2. Supercharge your knowledge base with a natural language Q&A platform! source: https://medium.com/@desmond2112/supercharge-your-knowledge-base-with-a-natural-language-q-a-platform-a56adc1fd0ba 3. The Era of Acute GPT Shortage Will Soon Begin source: https://tomaszurbanski.substack.com/p/the-era-of-acut..

1. LIMA: Less Is More for Alignment source: https://arxiv.org/pdf/2305.11206.pdf 2. Putting it all together: the implemented transformer source: https://medium.com/@hunter-j-phillips/putting-it-all-together-the-implemented-transformer-bfb11ac1ddfe

1. How does GPT-3 spend its 175B parameters? source: https://www.lesswrong.com/posts/3duR8CrvcHywrnhLo/how-does-gpt-3-spend-its-175b-parameters

1. Dr. LLaMA: Improving Small Language Models in Domain-Specific QA source: https://arxiv.org/pdf/2305.07804.pdf 2. GPT-4's Hidden Cost: Is Your Language Pricing You Out of AI Innovation? source: https://tomaszurbanski.substack.com/p/the-hidden-price-tag-on-gpt-4-for 3. Supercharge your knowledge base with a natural language Q&A platform! source: https://medium.com/@desmond2112/supercharge-y..