(오늘의 짤방: もしもAppleがユーザーの意見を商品に取り入れてたらどうなってたかw。(만약 애플이 사용자들의 의견을 제품에 반영했다면 어땠을지 궁금하다.) @tmiyatake1)
- 빅데이터/인공지능
- Announcing search.vespa.ai
Today, we announce the general availability of search.vespa.ai - a new search experience for all (almost) Vespa-related content - powered by Vespa, LangChain, and OpenAI’s chatGPT model.
- NVIDIA TensorRT-LLM Supercharges Large Language Model Inference on NVIDIA H100 GPUs
- Here's the most effective GPT-4 prompt I've developed for writing tasks.
- Baseline Defenses for Adversarial Attacks Against Aligned Language Models
- Elasticsearch Vector Store
- We just released Transformers.js v2.6.0! New features:
- 중국은 지금 LLM 전쟁 중··· 텐센트, 기업용 AI 모델 ‘훈위안’ 출시
- nanoT5 (Encoder-Decoder / Pre-training + Fine-Tuning)
- Large language models, explained with a minimum of math and jargon
- NAVER GLACE AI 개발팀은 어떤 일을 해?
- Why Large Language Models Hallucinate and How to Reduce it
- OpaquePrompts serves as a privacy layer around your LLM of choice.
- Open Interpreter lets LLMs run code on your computer to complete tasks.
- “생성형 AI로 합성 데이터 생성”··· AWS, 비접촉 결제 시스템 개발 사례 공유
- Hugging Face, Training Cluster As a Service 공개 (huggingface.co)
- Hierarchical Agents with LlamaIndex
- OnPrem.LLM is a simple Python package that makes it easier to run large language models (LLMs) on non-public or sensitive data and on machines with no internet connectivity (e.g., behind corporate firewalls).
- Ever wondered how LLMs generate text❓🤔
- 생성 AI 모델의 임베딩 벡터를 이용한 서버리스 추천 검색 구현하기 - 이상현 CEO, Mirror Inc. :: AWS Summit Seoul 2023
- Removing Demographic Data Can Make AI Discrimination Worse
- LlamaIndex Update — 09/03/2023
- 3 free MLOps courses you should know about:
- The ggml roadmap is progressing as expected with a lot of infrastructural development already completed
- 핵심 미래 분석 기술! 시계열 분석을 활용한 수요예측과 재고관리 최적화 사례 – 김형일 AWS 솔루션즈 아키텍트, 이환기 신세계아이앤씨 AI팀장:: AWS Cloud Week
- Falcon 180B 모델 공개 (huggingface.co)
- SAM.cpp - Meta의 Segment Anything Model을 순수 C/C++ 로 구현 (github.com/YavorGIvanov)
- Single Sample Training Experiment(Llama 2 7B + Platypus, using LoRA)
- AI Grant — accelerator for AI startups
- Welcome to Verba: The Golden RAGtriever, an open-source initiative designed to offer a streamlined, user-friendly interface for Retrieval-Augmented Generation (RAG) applications.
- Fine-Tuning LLMs: LoRA or Full-Parameter? An in-depth Analysis with Llama 2
- AI를 활용한 교육 (openai.com)
- SLiMe: Segment Like Me
- 놀이동산도 있네··· 구글, 클라우드 AI 기업 실사례 7가지 제시
- '오픈소스' AI에 대해 다시 생각해야 할 때
- ["쉿! 유출 안돼" 삼성, 반도체 개발에 '네이버 AI' 쓴다
- How to Use Streamlit and Python to Build a Data Science App
- DiffBIR: Towards Blind Image Restoration with Generative Diffusion Prior
- Fine-Tuning a Linear Adapter for Any Embedding Model
- ControlMat: A Controlled Generative Approach to Material Capture
- Syncing data sources to vector stores
- TSMixer: An all-MLP architecture for time series forecasting
- Code Llama: Open Foundation Models for Code
- KOTE (Korean Online That-gul Emotions) 데이터셋을, Haidt의 도덕감정(Moral Emotion) 분류로 편집한 멀티라벨(Multi-Labels) 데이터셋입니다. 해당 데이터셋을 KcELECTRA 훈련한 모델도 함께 공개합니다.
- Falcon-180B Demo
- Falcon 180B is a super-powerful language model with 180 billion parameters, trained on 3.5 trillion tokens.
- A complete guide to fine-tuning Code Llama
- AI 이미지 검색 엔진 만들기 - 벡터 데이터베이스 설명과 Chroma DB 튜토리얼
- 나를 가장 잘 이해하는 개인화 어시스턴트, LINER Copilot
- Trying to fine-tune an open-source LLM for your own data?
- 핑크퐁의 통합 데이터 환경 구축기 (feat. Snowflake)
- Shaq - CLI용 Shazam(음악 이름 찾기) (blog.yossarian.net)
- Copilots are becoming the new paradigm how to build successful LLM-based applications
- ‘AI 주권전쟁’ 승패 좌우할 LLM 경쟁력 ‘빨간불’
- 네이버 클로바 케어콜, 클라우드 서비스 시작
- 읽씹할 결심 - 생성 모델에게 답변 시간 가르치기
- 대규모 언어모델 너도 나도 업무에 사용해보자: To Beginner
- A new way to use computers - Open Interpreter lets LLMs run code on your computer to complete tasks.
- LangFuse - LLM앱을 위한 오픈소스 Observability & Analytics 솔루션 (github.com/langfuse)
- 구글 제미니가 오픈API GPT-4를 압도할까?
- LLM + Spreadsheets
- Google made a watermark for AI images that you can’t edit out
- LLM now provides tools for working with embeddings
- ♻️ Self-querying @zilliz_universe
+ DashVector
- Our new paper defines situational awareness for LLMs & shows that “out-of-context” reasoning improves with model size.
- CS480/680: INTRODUCTION TO MACHINE LEARNING, Spring 2023, University of Waterloo
- Whisper Web - ML-powered speech recognition directly in your browser
- Large language models in medicine: the potentials and pitfalls
- TokenMonster is an ungreedy subword tokenizer and vocabulary generator, enabling language models to run faster, cheaper, smarter and generate longer streams of text.
- Publicly Shareable Clinical Large Language Model Built on Synthetic Clinical Notes
- How to Use Large Language Models (LLMs) on Private Data: A Data Strategy Guide
- Large Content And Behavior Models To Understand, Simulate, And Optimize Content And Behavior
- GPT-3.5 Turbo 파인튜닝(colab)
- The Belebele Benchmark: a Parallel Reading Comprehension Dataset in 122 Language Variants
- 생성형 AI를 이용한 마케팅/SEO/PR용 자동화 스크립트 모음 (github.com/ktynski)
- Build generative AI applications with Google
- How to Create a Beautiful Polar Histogram With Python and Matplotlib
- Just implement the on_message function in main.py and Textbase will take care of the rest :) Since it is just Python you can use whatever models, libraries, vector databases and APIs you want.
- Meta, 컴퓨터 비전 모델의 편향성을 조사하기 위한 FACET 데이터셋 공개 (ai.meta.com)
- LangChain Library Adds Full Support for Neo4j Vector Index
- DSPy: Programming—not prompting—Foundation Models
- We write your reusable computer vision tools. Whether you need to load your dataset from your hard drive, draw detections on an image or video, or count how many detections are in a zone.
- Awesome Machine Learning for Cyber Security
- They propose applying the Segment Anything Model (SAM) to medical 2D images.
- A minimal study plan for Machine Learning 📈
- pdf2embeddings - NLP tool for scraping text from a corpus of PDF files, embedding the sentences in the text and finding semantically similar sentences to a given search query
- Building a PDF Knowledge Bot With Open-Source LLMs - A Step-by-Step Guide
- 한국은 AI 강국이 될 수 있을까?
- Start Machine Learning in 2023 - Become an expert for free!
- Transformers as Support Vector Machines
- Meta가 2023년 8월 31일 영상 처리 모델 "DINOv2"의 라이센스를 CC BY-NC 4.0에서 Apache License 2.0으로 변경
- Inside the AI Porn Marketplace Where Everything and Everyone Is for Sale
- How do domain-specific chatbots work? An Overview of Retrieval Augmented Generation (RAG)
- Interactive Tools for machine learning, deep learning, and math
- PyTorch Memory Tuning
- Time-series machine learning at scale
- CodeGeeX: A Multilingual Code Generation Model
- 의료계, AI 기반 진단과 치료 시행 기대
- Complete Numpy Library(YouTube)
- 구글 "언어 이해 시작한 AI, 헬스케어에도 혁신 가져올 것"
- Data Parallel Training with KerasNLP
- Build Your Own PandasAI with LlamaIndex
- All 161 things we announced at Google Cloud Next ‘23 – a recap
- GPT is becoming a Turing machine: Here are some ways to program it
- Comic-Factory - 디퓨전 모델을 이용하여 만화를 생성해주는 도구 (huggingface.co)
- AI: RAG vs Fine-tuning — Which Is the Best Tool to Boost Your LLM Application?
- 생성형 AI를 위해 지속적 테스트를 업그레이드하는 3가지 방법
- “아직 완성된 상태 아니다” 기업이 ‘생성형 AI’ 정책을 만들어야 하는 이유
- [대전환 AI시대⑬] "누구나 경험하는 금융 AI···수요 기반 전략은 필수"
- “사용자 대신해 회의 참석한다” 구글, 워크스페이스용 '듀엣 AI' 출시
- "AI 쓰는 직원 늘어나는데 관리 지침은 전무" 아사나 직장 내 AI 활용 설문조사
- 오픈AI, 기업용 챗GTP 출시··· “사용 제한 없고 속도 2배 더 빠르다”
- 구글, 듀엣 AI 공식 출시··· ‘기업용 워크스페이스에 도입, 월 30달러’
- 구글, 기업의 LLM 역량 강화 지원하는 버텍스AI 새 기능 공개
- 엔비디아-구글 클라우드, 파트너십 확대… “새로운 AI 인프라와 소프트웨어 출시”
- AI가 직면한 중대한 문제점 “대중이 신뢰하지 않는다”
- PromptTools - Welcome to prompttools created by Hegel AI! This repo offers a set of open-source, self-hostable tools for experimenting with, testing, and evaluating LLMs, vector databases, and prompts.
- Quantitative Evaluation of LLM Responses with RAG-based Question-answering Chatbots
- Embrace the randomness
- Google - Fast or Slow? Predict AI Model Runtime
Predict how fast an AI model runs
- AI and Data Scientist Roadmap
- Artificial Intelligence - Creating The Assembly Line For Knowledge Workers
- When Should You Stop Searching? - An introduction to optimal stopping and how it relates to data science
- This notebook walks through using Marvin to extract and augment metadata from text. Marvin uses the LLM to identify and extract metadata.
- theres been a lot of excitement around fine-tuning recently, both in open source and with OpenAI's API - Here’s a list of some of our favorite resources, use-cases, and experiments on the topic over the last ~week
- Automatic Generation of Visualizations and Infographics with LLMs
- We’re All Programmers Now
With generative AI, anyone can code. Here’s how to help your enterprise embrace this change.
- Awesome Llama2 Resources
- Don’t re-record, Regenerate. How? AI-powered generative audio technology instantly recreates any voice and completely matches its recording environment, making your edits sound seamless and natural.
- 삼성, AI기반의 개인화된 음식 & 레시피 서비스 '삼성 푸드' 공개 (news.samsung.com)
- Optimize open LLMs using GPTQ and Hugging Face Optimum
- One Project To Learn MLOps
- DeepSpeed-Chat: Llama/Llama-2 system support, efficiency boost, and training stability improvements
- Microsoft LIDA - LLM을 이용한 시각화/인포그래픽 자동 생성 (microsoft.github.io)
- There's an amazingly convenient way to install the *full* NVIDIA CUDA dev stack on Linux.
- Why LLMs kickass at Code Generation and will they Replace Programmers?
- Accelerating PyTorch with CUDA Graphs
- Vector Search with OpenAI Embeddings: Lucene Is All You Need
- DeepSpeed Ulysses: 긴 시퀀스 트랜스포머 모델 훈련을 위한 시스템 최적화 (github.com/microsoft)
- Awesome-LLM - 🔥 Large Language Models(LLM) have taken the NLP community AI community the Whole World by storm. Here is a curated list of papers about large language models, especially relating to ChatGPT.
- 오픈AI와 자체구축 LLM을 비교하다
- SKT, 기업용 LLM 시장 나선다..."에이닷·앤트로픽·코난 3개 모델 조합"
- 6 ways machine learning can boost your marketing processes
- 유진투자증권, 두물머리와 챗GPT 활용한 'AI 애널리스트 솔루션' 선봬
- GN⁺: 메타 AI, 비디오 상의 모든 점(픽셀)을 추적하는 모델인 CoTracker를 발표 (co-tracker.github.io)
- 모자이크ML, 오라클 클라우드 인프라스트럭처 활용해 생성형 AI 모델 학습 속도 개선한다
- Considerations for Deploying Machine Learning Models in Production
- DeepEval provides a Pythonic way to run offline evaluations on your LLM pipelines so you can launch comfortably into production.
- Building Performant RAG Applications for Production
- LLM과 함께 뜨는 중··· 개발자를 위한 '랭체인' 안내서
- Code Llama is a family of large language models for code based on Llama 2 providing state-of-the-art performance among open models, infilling capabilities, support for large input contexts, and zero-shot instruction following ability for programming tasks.
- VisionScript is an abstract programming language for doing common computer vision tasks, fast.
- Awesome Systematic Trading - We are collecting a list of resources papers, softwares, books, articles for finding, developing, and running systematic trading (quantitative trading) strategies.
- Machine Learning Recipes
- LlamaIndex: Automatic Knowledge Transfer (KT) Generation for Code Bases
- LLMs in real-world projects
- Shell-AI: Your Intelligent Command-Line Companion
- Reinventing the LSTM: Long short-term memory from scratch
- vectorizer.ai - Trace Pixels To Vectors in Full Color
- MedAlign: A Clinician-Generated Dataset for Instruction Following with Electronic Medical Records
- ecoding ROC Analysis
- Web LLM - Llama2 7B/13B 지원 시작 (webllm.mlc.ai)
- The state of AI in 2023: Generative AI’s breakout year
- Containerized PDF Summarizer with FastAPI and Hamilton
- MS-Vid2Vid-XL aims to improve the spatiotemporal continuity and resolution of video generation.
- A key piece of advice I give to devs on LLM + RAG evals is to evaluate piece-by-piece: e.g. evaluate the retrieval step on its own.
- Llama2가 요약에 있어 GPT-4만큼 정확하며 30배 더 저렴 (anyscale.com)
- Xata x LangChain: new vector store and memory store integrations
- DeepEval is a cool library for testing your LLM and RAG apps
- Nougat: Neural Optical Understanding for Academic Documents
- CppCon 2016: “Bringing Clang and C++ to GPUs: An Open-Source, CUDA-Compatible GPU C++ Compiler"
- Scikit-LLM: Sklearn Meets Large Language Models
- WizardCoder: Empowering Code Large Language Models with Evol-Instruct
- Beating GPT-4 on HumanEval with a Fine-Tuned CodeLlama-34B
- Welcome to Google Cloud Next ’23
- 4 Charts That Show Why AI Progress Is Unlikely to Slow Down
- Newspaper3k Guide: Scrape Articles Using AI
- How AI Is Changing the Way We Code
- Looking to easily run Code Llama / Llama-2 on your local machine? 🚀
- 코딩에 특화된 LLM··· 메타, 코드 라마(Code LIama) 공개
- "챗GPT 언급한 채용 공고, 21배 증가"··· 기업들의 동향은?
- ‘규제 불확실성, 데이터 거버넌스...’ 생성형 AI의 현 과제는? 해결 전략은?
- This repository is a curated collection of links to various courses and resources about Artificial Intelligence (AI). Whether you're a beginner or an experienced learner, there's something here for everyone!
- Data Observability for Analytics and ML teams
- In this video you will learn to create a Langchain App to chat with multiple PDF files using the ChatGPT API and Huggingface Language Models.
- Onboard - GitHub Repo 설명해주는 AI (getonboard.dev)
- Calculate and visualize derivatives in Python
- ChatGPT 엔터프라이즈를 소개합니다. (openai.com)
- Introducing ChatGPT Enterprise
- Parameter-Efficient Fine Tuning (PEFT) methods freeze the pretrained model parameters during fine-tuning and add a small number of trainable parameters (the adapters) on top of it.
- How to deploy ML models painlessly
- LlamaIndex Finetuning
Overview
- Estimating vRAM - Determining if you have enough GPU memory.
- 야심차게 출발했지만 … 네이버AI, 온종일 대기중
- Tensorflow Object Detection in 5 Hours with Python | Full Course with 3 Projects
- Master Artificial Intelligence in 2023 with these free courses:
- LLM 기반의 시스템 & 제품 구축 패턴들 (eugeneyan.com)
- Graph LLM, Demo Outline
- Google Gemini Eats The World – Gemini Smashes GPT-4 By 5X, The GPU-Poors
- Demo: CodeLlama-13b with MLC LLM
- 👋 Welcome to MLC LLM¶
- Embedding Playground
- 네이버 '클로바X' 내놓자 삼성도 AI 출사표
- 10월 뉴로클라우드 출시…네이버클라우드, 기업 AI 자신
- TIL CUDA_VISIBLE_DEVICES=0 won't always give you the same GPU, unless you also set CUDA_DEVICE_ORDER=PCI_BUS_ID
- **timeline** based data-structure to do RAG. Really useful for querying over personal events
- The unstructured library provides open-source components for ingesting and pre-processing images and text documents, such as PDFs, HTML, Word docs, and many more. The use cases of unstructured revolve around streamlining and optimizing the data processing workflow for LLMs.
- 실시간 채팅 번역에 최적화된 프롬프트 만들기
- Generative AI and intellectual property
- Efficient semantic search over unstructured text in Neo4j
- This might be the first time ChatGPT (+@jxnlco) helped us come up with a better retrieval algorithm for RAG:
- 한국판 챗GPT 네이버 클로바X 사용 후기(성능 테스트, 스킬 기능 소개)
- ChatGPT를 임상현장에서 Co-pilot처럼 사용할 수 있는지 연구 결과 소개
- 금융도메인에서의 생성형 언어모델의 역할과 미래
- 네이버 챗봇 '클로바X' 써보니…느리지만 맛집·길찾기·면접 '제법 똑똑하네'
- Reverse Engineering Image Prompts with PromptPerfect
- REVEALED: THE AUTHORS WHOSE PIRATED BOOKS ARE POWERING GENERATIVE AI
- Leveraging XGBoost for Time-Series Forecasting
- 하드웨어
- 읽을거리
(보너스: everything is an API call. free your mind. via @_nateraw)
EOB
댓글 없음:
댓글 쓰기