[GitHub](https://github.com/aiagenta2z/awesome-chatgpt-apps) | [Website](https://www.deepnlp.org/store/chatgpt-app)In this blog, I will introduce you step by step on how to submit your First ChatGPT Apps.ChatGPT Apps are just like the Apple App Store and WeChat Mini Apps, in which developers implement their features as 'tool' and UIas "resources", and render back the results to ChatGPT and display in the Chat Interface. In this blog, we will use a simple Bill Agentwhich help you track daily bills and generate expenses summary report as an example of how to submit your AI Apps/Agents to ChatGPT.Required:0. App Icon Format: SVG (64 × 64 px)1. A Valid MCP Server URL: This is the core part of an ChatGPT Apps, you can test if the tools are resources are rendered correcly. I tried cursor, DeepNLP Agent Router(https://www.agent.deepnlp.org)2. Domain Verification. You will need to put a verification code under your root domain to verify ownership, the address is 3. 5 Test Cases and 3 negative cases: Which Allow ChatGPT to know when to call and not to call your tools.4. Screenshot Sizes: ### submit address https://platform.openai.com/apps-manage### guideline: https://developers.openai.com/apps-sdk/app-submission-guidelinesWe have wrapped up a simple add bill payment agent which you can add a bill by typing something like "xxxx"https://agent.deepnlp.org/container/aiagenta2z/a2z_bill_agent/mcpStep 1. App InfoThis is the place where you add your App name, Subtitle, Description of your App and choose category.Step 2. MCP Server This is the place where you let ChatGPT to connect to your MCP server and get tools/ resources. You also need to provide some information about the MCP field justification such as why the tool is not read only (In our case, we are adding a bill record to MCP server DB), etc.Domain VerificationIn this like the Google search console, you have to put the code under the relative path so that website can read the file and know you own the domain. https://{your_domain}/.well-known/openai-apps-challengeStep 3. TestingThis is the step which takes the most time to complete.In this step, you need to fill in form of providing 5 positive cases and 3 negative cases to ChatGPT so that they can use it as system prompts to let ChatGPT properly chose tools. Step 4. ScreenShotsThis step is easy, just prepare 4 images of the required sizes.Changes these 3 files to At least 1 screenshot is required at 2x (retina-quality); 4 screenshots max.Requirements:PNG fileWidth: 706pxMin height: 400pxMax height: 860px (Recommended)Step 5. GlobalChoose CoutriesStep 6. Submit Submit and consend to a lot of options, easy steps.

Hi, I am a developer who have some basic knowledge of AI Agent, RAG and Multi-Agent Dialogues. Now I have a small project on hand, which I need to build a AI Agent on the finance industry to give some realtime information to my clients. When I am choosing different AI Agent platforms, I have some difficulties. Right now, I am comparing among Google Vertex AI Agent Builder, Microsoft Azure AI Agents and Salesforce AI Agents. Any suggestions or some free AI Agent Builder recommendations?I am actually cost-sensitive, for example, Google Vertex AI Agent builders have price is $12 per 1,000 queries, and Vertex AI search is $2 per 1,000 queries, which is a little bit above my budget limit.

[GitHub](https://github.com/aiagenta2z/awesome-chatgpt-apps) | [Website](https://www.deepnlp.org/store/chatgpt-app)In this blog, I will introduce you step by step on how to submit your First ChatGPT Apps.ChatGPT Apps are just like the Apple App Store and WeChat Mini Apps, in which developers implement their features as 'tool' and UIas "resources", and render back the results to ChatGPT and display in the Chat Interface. In this blog, we will use a simple Bill Agentwhich help you track daily bills and generate expenses summary report as an example of how to submit your AI Apps/Agents to ChatGPT.Required:0. App Icon Format: SVG (64 × 64 px)1. A Valid MCP Server URL: This is the core part of an ChatGPT Apps, you can test if the tools are resources are rendered correcly. I tried cursor, DeepNLP Agent Router(https://www.agent.deepnlp.org)2. Domain Verification. You will need to put a verification code under your root domain to verify ownership, the address is 3. 5 Test Cases and 3 negative cases: Which Allow ChatGPT to know when to call and not to call your tools.4. Screenshot Sizes: ### submit address https://platform.openai.com/apps-manage### guideline: https://developers.openai.com/apps-sdk/app-submission-guidelinesWe have wrapped up a simple add bill payment agent which you can add a bill by typing something like "xxxx"https://agent.deepnlp.org/container/aiagenta2z/a2z_bill_agent/mcpStep 1. App InfoThis is the place where you add your App name, Subtitle, Description of your App and choose category.Step 2. MCP Server This is the place where you let ChatGPT to connect to your MCP server and get tools/ resources. You also need to provide some information about the MCP field justification such as why the tool is not read only (In our case, we are adding a bill record to MCP server DB), etc.Domain VerificationIn this like the Google search console, you have to put the code under the relative path so that website can read the file and know you own the domain. https://{your_domain}/.well-known/openai-apps-challengeStep 3. TestingThis is the step which takes the most time to complete.In this step, you need to fill in form of providing 5 positive cases and 3 negative cases to ChatGPT so that they can use it as system prompts to let ChatGPT properly chose tools. Step 4. ScreenShotsThis step is easy, just prepare 4 images of the required sizes.Changes these 3 files to At least 1 screenshot is required at 2x (retina-quality); 4 screenshots max.Requirements:PNG fileWidth: 706pxMin height: 400pxMax height: 860px (Recommended)Step 5. GlobalChoose CoutriesStep 6. Submit Submit and consend to a lot of options, easy steps.

大家在使用拼多多(PPD Temu)APP的搜索推荐Search and Recommendation 功能的时候遇到了哪些好的体验和有问题的体验？请麻烦写明复现条件，比如prompt输入文本，上传截图。

大家在使用支付宝(Alipay)APP的搜索推荐Search and Recommendation 功能的时候遇到了哪些好的体验和有问题的体验？请麻烦写明复现条件，比如prompt输入文本，上传截图。

大家在使用淘宝(Taobao)APP的搜索推荐Search and Recommendation 功能的时候遇到了哪些好的体验和有问题的体验？请麻烦写明复现条件，比如prompt输入文本，上传截图。

大家在使用京东(JD)APP的搜索推荐Search and Recommendation 功能的时候遇到了哪些好的体验和有问题的体验？请麻烦写明复现条件，比如prompt输入文本，上传截图。

大家在使用知乎(Zhihu)APP的搜索推荐Search and Recommendation 功能的时候遇到了哪些好的体验和有问题的体验？请麻烦写明复现条件，比如prompt输入文本，上传截图。

大家在使用微信(WeChat)APP的搜索推荐Search and Recommendation 功能的时候遇到了哪些好的体验和有问题的体验？请麻烦写明复现条件，比如prompt输入文本，上传截图。

大家在使用小红书(Xiaohongshu)APP的搜索推荐Search and Recommendation 功能的时候遇到了哪些好的体验和有问题的体验？请麻烦写明复现条件，比如prompt输入文本，上传截图。

大家在使用快手(Kuaishou Kwai)短视频的搜索推荐Search and Recommendation 功能的时候遇到了哪些好的体验和有问题的体验？请麻烦写明复现条件，比如prompt输入文本，上传截图。

大家在使用抖音视频的搜索推荐Search and Recommendation 功能的时候遇到了哪些好的体验和有问题的体验？请麻烦写明复现条件，比如prompt输入文本，上传截图。

In this blog, we will give you a brief introduction of most recent progress in On-Device Recommendation (Edge Recommendation) in real-world applications. Mobile AI systems and applications have been more popular due to increasing number of mobile devices and technology developments in deep learning based methods, e.g. model compression, distillation and so on. In recent years, on-device recommendations have enpowered many Mobile Apps to better respond to users' most real-time behaviors on mobile deivces, including clicks, scroll-donwns, likes, and many others. We will introduce three applications, including EdgeRec in Taobao, searchbar background words reranking in Alipay, search result reranking in Meituan-Dianping, short-video recommendation in KuaiShou, TfLite Implementation of Tensorflow, etc.

Ranking is a crucial part of modern commercial recommendation and advertisement system. It aims to solve the problem of accurate click-through rate(CTR) prediction. In this article, we will provides some of most popular ranking equations of commercial recommendation or ads system.

This blog summarizes the latest research development of Retrieval papers published in KDD2023 conferences. This year there are total 8 papers related to Retrieval in KDD2023. Most of the authors' affiliations are top research institutes (Google Research, DeepMind, Meta FAIR) and universities (Stanford, Berkeley, MIT, CMU and others).

This blog summarizes the latest research development of Recommendation papers published in KDD2023 conferences. This year there are total 44 papers related to Recommendation in KDD2023. Most of the authors' affiliations are top research institutes (Google Research, DeepMind, Meta FAIR) and universities (Stanford, Berkeley, MIT, CMU and others).

Cross-Domain Recommendation in Commercial Recommendation System With application of MMD and Wasserstein distance

In this post, we will talk about some real-world applications of deep candidates generation (vector-retrieval) models in the matching stage of recommendation scenario. Commercial recommendation system will recommend tens of millions of items to each user. And the recommendation process usually consists of two stages: The first stage is the candidate generation(matching) stage, a few hundred candidates are selected from the pool of all candidate items. The second stage is the ranking stage in which hundreds of items are ranked and sorted by the ranking score. Then the top rated items are displayed to users.