AI Knowledge Base

Author: Han Fangyuan, author of "Dify on WeChat" open source project 1. Overview WeChat, as the most popular instant messaging software, has a huge amount of traffic. WeChat's friendly chat window is a natural AI application LUI (Language User Int...

Researchers have investigated a "jailbreak attack" technique - a method that can be used to bypass security fences set up by developers of large language models (LLMs). The technique, known as a "multisample jailbreak attack," uses Anthropic's own models, as well as those produced by other AI companies, to...

Original: https://arxiv.org/pdf/2210.03629.pdf Can't understand how ReAct works and applies even after reading it? See "ReAct Implementation Logic in Action" for an explanation with real-world examples. Abstract Although large languages ...

RAG (Retrieve Augmented Generation) is a technique for optimizing the output of large language models (LLMs) based on authoritative knowledge base information. This technique generates responses by extending the functionality of LLMs to...

ChatOllama is an open source chatbot based on LLMs. For a detailed description of ChatOllama click on the link below. ChatOllama | Ollama based 100% local RAG application ChatOl...

原文：《Dense X Retrieval: What Retrieval Granularity Should We Use?》 Note: This method is suitable for a small number of models, such as the OPENAI series, Claude series, Mi...

Original article: https://blog.langchain.dev/langfriend/?t=ZMdNze4qQKvNzUMHPhQRfg LangChain is very much focused on the concept of memory, and we're used to creating sample apps...

Langchain Chatchat is a knowledge base enhancement solution that enables: fully localized reasoning, focusing on solving the enterprise pain points of data security protection and private domain deployment. The open source solution is licensed under the Apache License and can be commercialized free of charge without paying ...

Today I read an article on Hacker News about a new plugin for Obsidian that integrates with ChatGPT. There are quite a few of these tools on the market and I'm happy to see the different ways people are applying them to Obsidian.They can help people connect...

Cue word analysis 1. With the help of XML tags, text cues are given a clear structure, which allows big language models to understand your instructions better. By the way, Claude 3 is very XML-friendly, and most of the high-quality cue word outputs are organized based on XML...

I read an interesting paper today, Large Language Models as Analogical Reasoners, that mentions a new approach to Prompt - the "analogical prompting method. If you are familiar with Prompt Engineering...

Original: Generally Capable Agents in Open-Ended Worlds [S62816] 1. Reflective Intelligents Able to inspect and modify self-generated code or content, and continually iteratively optimize it by self...

Step 1: Connect to a data source Visit the [My Files] tab and click the "Add File" button in the center of the screen. Select Upload File to upload a file in any format for Julius to analyze or edit, be it a CSV, Excel, JSON file...

This AI assistant is an expert in deep learning, transformers, diffusion models, and LLM development, specializing in Python libraries like PyTorch, Diffusers, Transformers, and Gradio. Here's what you can do with this helper...

Please use models that support long memory for writing, such as KIMI, Claude, etc. Because in the fourth step you need to introduce at least 3-5 pieces of content from the search results for reference and creation. During the creation process the output may be interrupted due to the length of the input text, the above two models by input...

Abstract The reasoning performance of Large Language Models (LLMs) on a wide range of problems relies heavily on chained-thinking cues, which involves providing some chained-thinking demonstrations as examples in the cues. Recent research, e.g., thinking trees, has pointed to exploration and self-assessment of reasoning steps in complex problem solving ...

In Natural Language Processing (NLP), intent refers to a user's expression of some purpose, want or desire. By analyzing the messages sent by the user and recognizing the intention behind them, we can reply with relevant content. For example, "order food", "check the weather", "I want to go to Paris", etc. are valid intents. To...

English You are a sophisticated researcher and professor Ask for a dataset and a field. When it is uploaded...

Organize Long Text Prompt: "The following is a transcript of an interview with YYY by XXX that I have put together, please organize it into a more readable conversational format, form chapters based on the topic, include names of people, embellish appropriately while respecting the original intent, and don't leave anything out from the beginning!" As...

TL;DR This article is 8200+ words and takes about 15 minutes to read in full. This article briefly reviews the history of ChatGPT, from perceptual machines to the latest big model application for deep learning. Original article: https://hutusi.com/articles/the-hi...

1. Key principles Conciseness and technicality: Write accurate and concise technical responses, providing Python examples. Readability and Repeatability: Ensure that the data analysis process is readable and easily reproducible by others. Functional Programming: Use functional programming where appropriate and try to avoid unnecessary...

Source URL: https://chat.openai.com/g/g-n7Rs0IK86-grimoire Note: The following is an old version of the original article, to get the latest version of the prompt word, you need to log in to your ChatGPT member account and use the tips provided...

What is Page Setup and how do I access it? Page Setup is a tab in Gamma that allows you to make advanced changes to Gamma and how it is configured. You can perform a variety of actions, from changing the background of Gamma to configuring its Search Engine Optimization (SEO) meta...

How to Import PPT, Google Slides, Google Docs and Word Documents To get started with Gamma, you can extract documents or presentations created in other tools. Currently, we support importing from Google Docs, Goog...

Tweet tone detector Original System: Your task is to analyze the provided tweet and identify the primary...

Spreadsheet sorcerer Original System: Your task is to generate a CSV spreadsheet containing the specifie...

Product naming pro Original System: Your task is to generate creative, memorable, and marketable product...

Polyglot superpowers Original System: You are a highly skilled translator with expertise in many languag...

Socratic sage Original System: You are an AI assistant capable of having in-depth Socratic style convers...

Lesson planner Original System: Your task is to create a comprehensive, engaging, and well-structured le...

Email extractor Original System: You are an AI assistant with expertise in LaTeX, a document preparation...

Motivational muse Original System: Your task is to generate a personalized motivational message or affir...

Direction decoder Original System: Your task is to take the provided natural language description of a p...

Review classifier Original System: You are an AI assistant trained to categorize user feedback into pred...

Efficiency estimator Original System: Your task is to analyze the provided function or algorithm and cal...

Brand builder Original System: Your task is to create a comprehensive design brief for a holistic brand ...

Data organizer Original Text System: Your task is to take the unstructured text provided and convert it into ...

Code clarifier Original System: Your task is to take the code snippet provided and explain it in simple...

Riddle me this original article Generate a clever riddle and provide a step-by-step guide to help the user arriv...

Grammar genie Original System: Your task is to take the text provided and rewrite it into a clear, gramm...

Interview question crafter Original System: Your task is to generate a series of thoughtful, open-ended ...

Grading guru 原文 Evaluate the following two texts based on the given criteria: Text 1: The sun was ...

Career coach Original System: You will be acting as an AI career coach named Joe created by the company ...

Memo maestro Original System: Your task is to compose a comprehensive company memo based on the provided...

PII purifier Original System: You are an expert redactor. The user is going to provide you with some tex...

Claude 3 System Alert Original The assistant is Claude, created by Anthropic. The current date is March 4th, 2024...

Second-grade simplifier Original System: Your task is to take the text provided and rewrite it in a way ...

Mindfulness mentor Original System: You are an AI assistant with expertise in mindfulness and stress man...

Trivia generator Original text Generate trivia questions on various topics and provide hints to help users ar...

Perspectives ponderer Original Article Analyze the pros and cons of implementing a four-day workweek as a standa...

Prose polisher Original System: You are an AI copyeditor with a keen eye for detail and a deep understan...

Emoji encoder Original System: Your task is to take the plain text message provided and convert it into ...

CSV converter Original System: As a data conversion expert, your task is to convert data from different ...

Neologism creator Original text Can you help me create a new word for the act of pretending to understand som...

Function fabricator Original System: Your task is to create Python functions based on the provided natur...

Code consultant Original System: Your task is to analyze the provided Python code snippet and suggest im...

Idiom illuminator Original System: Your task is to provide a clear explanation of the meaning and origin...

Meeting scribe Original System: Your task is to review the provided meeting notes and create a concise s...

Ethical dilemma navigator Original System: Help the user navigate a complex ethical dilemma by identi...

Simile savant Original text Help me create some similes to describe a person's laughter that is joyful and co...

Git gud Original article I have made some changes to my local files and I want to save them in my local Git repo...

Mood colorizer Original System: Your task is to take the provided text description of a mood or emotion ...

LaTeX legend Original System: You are an AI assistant with expertise in LaTeX, a document preparation sy...

Hal the humorous helper Original System: You will play the role of Hal, a highly knowledgeable AI assist...

Portmanteau poet Original System: You are an AI assistant with a knack for creating innovative portmante...

Culinary creator Original System: Your task is to generate personalized recipe ideas based on the user's...

Pun-dit Original System: You are an AI assistant with a witty sense of humor and a knack for crafting cl...

Dream interpreter Original System: You are an AI assistant with a deep understanding of dream interpreta...

SQL sorcerer Original System: Transform the following natural language requests into valid SQL queries. ...

Cite your sources Original System: You are an expert research assistant. Here is a document you will ans...

Storytelling sidekick Original System: You are an AI assistant with a passion for creative writing and s...

Time travel consultant Original System: You are an AI assistant with expertise in physics, philosophy, a...

Python bug buster Original System: Your task is to analyze the provided Python code snippet, identify an...

Google apps scripter Original article Write me a Google apps script that will translate all text in a Google Sli...

Excel formula expert Original System: As an Excel Formula Expert, your task is to provide advanced Excel...

Website wizard Original System: Your task is to create a one-page website based on the given specificati...

Corporate clairvoyant Original text Your task is to analyze the following report: [Full text of Matterport SEC...

Cosmic keystrokes Original article Write me a fully complete web app as a single HTML file. The app should conta...

1.Bind domain name in cloudflare 2.cloudflare Tunnels Deploy on server as per tutorial and set forwarding port and domain name for the website 3.Bind 172.0.0.1:port (nginx) to server web directory ...

Overview Live Generation is an addition to our family of real-time features on the Leonardo.Ai platform. It lets you instantly see images being generated and changed as you type text. You can constantly restate, rewrite and refine your ideas. About real-time generation Refinement...

Overview Live Canvas is the result of the latest image conversion technology that we have adopted. This feature quickly transforms your drawings into stunning works of art, revolutionizing the way users interact with AI to generate images. It provides an immersive and intuitive creative experience, like painting digitally...

It's useful to keep AI-generated character designs consistent across multiple scenarios, whether you're designing a cover for a book series or producing concept art. However, it's no easy task - especially when using an illustration style. To simplify the process, we've crafted this guide that covers what you need to do to create consistent character...

Introduction Training your own image generation models using Leonardo.Ai can be a highly fulfilling experience once you have mastered an effective method. Model fine-tuning, which is its formal name, is a powerful feature that allows you to customize the output to your stylistic needs. In the tour...

Canvas Editor is a powerful image editing tool that lets you make fine adjustments to your images, remove unwanted elements, and even create new content using advanced image repair and expansion techniques. With the [new release of Canvas Editor], you can now have more detailed control over image modifications...

After creating a beautiful photo, you may need to enlarge it for printing or make a background-free version for compositing. In the web application, you can directly utilize our AI post-processing tools to enlarge the image. 1. Use Alchemy to zoom in Visit [A...

In the process of creating images, even with the use of sophisticated text-to-picture hints, it may sometimes be necessary to fine-tune some details, as the image generation model is inherently randomized. To adjust the generated image, you can use the canvas editor. 1. Enter the Canvas Editor After generating the image...

Getting Started Tip The prompt input box is located at the top of the screen. Here you can enter exactly what you want to generate. You can enter just the elements you want to see, or you can enter complete sentences - as the system understands both very well. But the key is clarity - the more our AI tools can...

When you first tried to create an image in [Image Generation via Text Prompts], you probably started with a simple prompt such as "An orange cat by the window sill". This guide will draw on the wisdom of the Leonardo.Ai community, our hint generation feature, and some best practices for building hints to help...

When using Leonardo.Ai for image creation, the various features have different token consumption depending on their complexity and the computational resources required. Below is a clear table that lists the cost of tokens per use for each feature to help you better plan and control your usage. Function Single...

Introduction Welcome to the world of Leonardo Alchemy! This advanced tool brings unprecedented detail and control to your creative process. Enhance your designs with dramatically improved high resolution, contrast enhancement, resonance, and more. Whether you're new to creating or...

Welcome to Leonardo.AI! It's a pleasure to meet you. This guide will walk you through how to create your first images on our pioneering AI innovation platform. Create visuals, whether they are illustrations or marketing visuals, with unrivaled quality, fast speed and style of a...

All of these bots are server bots built on the API. You can try each of them in Poe by clicking on the links. Web Search This bot performs web searches and then writes answers to information provided by searches through the bot query API using GPT-3.5...

How to Create a Prompt Bot for Poe Step 1:Open the Create a Bot page If you are using a web browser, go to poe.com and click the "Create a bot" button on the left sidebar. You can also go directly to poe.com/creat...

Original article: [State-of-the-art Code Generation with AlphaCodium - From Prompt Engineering to Flow Engineering...

Chirp may sing slurred words at certain times.' Phantom hearing' timing is when the song deviates from the expected lyrics, or Chirp is unable to find lyrics that match the melody. We can't 'fix' the lyrics after the audio clip has been generated, but we try as much as possible to prevent the same when the audio is generated again...

Suno Metatags were utilized to guide the flow of lyrics and musical interludes, helping Chirp transition smoothly through the various parts of the song. Particular attention was paid to some of the metatags during the training phase of the AI model, but we also discovered many others! The style of music you choose and the song you create...

Hello everyone, today we're going to explore the technique of participles in large-scale language modeling (LLM). Unfortunately, disambiguation is a more complex and tricky part of current top LLMs, but understanding some of its details is very necessary because many people blame some of the shortcomings of LLMs on neural networks or other seemingly...

Thesis: https://arxiv.org/abs/2402.14207 Can we teach LLMs to write long articles from scratch, based on reliable sources? Do Wikipedia editors think this will help them? 📣 Announcing STORM, an Internet-based...

Browse AI allows you to extract and monitor data from any web page in minutes without code. The bots you train to operate websites just like humans do will allow you to extract the most accurate data with the least amount of effort. This guide introduces some key concepts to help you get started. ...