Knowledge Centre
CustomGPT
Resources: CustomGPT uses no-code visual builder is easy to use, even non-technical people have built amazing custom GPT chatbots.
- Quickstart Guide: Build in second by uploading your documents or website content. Quickstart Guide [email.customgpt.ai]
- ย Need help? Find free, educational (and entertaining) guides, videos and lessons on theย Documentation [email.customgpt.ai] portal.
- ย Need inspiration? See our growing library of Case studies [email.customgpt.ai].
Two ways to get started: Bring your AI vision to life without writing any code. Get started by uploading documents OR use website content, and use our easy no-code visual builder to build your custom GPT chatbot.
- Build using documents: Start by uploading some documents and get a custom chatbot in seconds. Just select "Create Project" and then the "Upload" tab to upload documents. We support 1400+ document formats.
- Build using website content: If you need to ingest website content, just input a sitemap into the "Sitemap" tab. Use our free tools to find your sitemap or create a custom sitemap from various forms of web content (websites, helpdesks, Youtube videos, podcasts, RSS feeds, Google results, and more)
- Create your first project [email.customgpt.ai].
Email CustomGPT here.
AI courses
๐๐ฒ๐ด๐ถ๐ป๐ป๐ฒ๐ฟ:
๐ญ. Introduction to AI - IBM: https://lnkd.in/eCXRJSmM
๐ฎ. AI Introduction by Harvard: https://lnkd.in/eNJ_4Cnp
๐ฏ. Intro to Generative AI: https://lnkd.in/e3fQzfFY
๐ฐ. Prompt Engineering Intro: https://lnkd.in/e3Ww6pDz
๐ฑ. Google's Ethical AI: https://lnkd.in/enpfVNCw
๐๐ป๐๐ฒ๐ฟ๐บ๐ฒ๐ฑ๐ถ๐ฎ๐๐ฒ:
๐ฒ. Harvard Data Science & ML: https://lnkd.in/ev_zVCPp
๐ณ. ML with Python - IBM: https://lnkd.in/eb2eMYt9
๐ด. Tensorflow Google Cloud: https://lnkd.in/eMAyvDbe
๐ต. Structuring ML Projects: https://lnkd.in/ehx8gP8W
๐๐ฑ๐๐ฎ๐ป๐ฐ๐ฒ๐ฑ:
๐ญ๐ฌ. Prompt Engineering Pro: https://learnprompting.org
๐ญ๐ญ. Advanced ML - Google: https://lnkd.in/eTCrgUBe
๐ญ๐ฎ. Advanced Algos - Stanford: https://lnkd.in/eqD39TVs
๐ ๐๐ผ๐ป๐๐:
Amazon's AI Strategy: https://lnkd.in/e39JKBk2
Python
- Python for Everybody Specialization, Charles Russell Severance (University of Michigan)
- Python Programming Fundamentals, Andrew D. Hilton, Genevieve M. Lipp, Nick Eubank, Kyle Bradbury (Duke University)
- Python for Beginners (Coursera)
- Introduction to Python for Econometrics, Statistics and Numerical Analysis, Kevin Sheppard
Chain-of-thought (CoT) Prompting for LLMs
Manual Chain-of-thought (CoT), the first paper that introduced CoT:ย
Zero-Shot CoT:ย ย
Different variations:
- Self-Consistency Improves Chain of Thought Reasoning in Language Models [arxiv.org]
- Self-Polish: Enhance Reasoning in Large Language Models via Problem Refinement [aclanthology.org]
- Dynamic Voting for Efficient Reasoning in Large Language Models [aclanthology.org]
- Making Language Models Better Reasoners with Step-Aware Verifier [aclanthology.org]
- Automatic Chain of Thought Prompting in Large Language Models [arxiv.org]
- Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models [aclanthology.org]
More advanced variations:
- Tree of Thoughts: Deliberate Problem Solving with Large Language Models [arxiv.org]
- Graph of Thoughts: Solving Elaborate Problems with Large Language Models [arxiv.org]
Some surveys worth checking:
LLM: Hallucinations and Annotation Capabilities
Hallucinations in LLMs
- Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models [arxiv.org] [Zhang et al_2023]
- HaluEval: A Large-Scale Hallucination Evaluation Benchmark for Large Language Models [aclanthology.org] [Li et al. 2023]
- Chain-of-Verification Reduces Hallucination in Large Language Models [Dhuliawala et al. 2023]
- MaxMin-RLHF: Towards Equitable Alignment of Large Language Models with Diverse Human Preferences [Chakraborty et al.]
Prompt-engineering for vision language models
- What does CLIP know about a red circle? Visual prompt engineering for VLMs [openaccess.thecvf.com] [Shtedritski_2023]
Annotation Capabilities of Large Language Models
- Machine-assisted mixed methods: augmenting humanities and social sciences with artificial intelligence [arxiv.org]
- Last Words: Empiricism Is Not a Matter of Faith [aclanthology.org]
- AFaCTA: Assisting the Annotation of Factual Claim Detection with Reliable LLM Annotators [arxiv.org]
- Isย ChatGPTย a Good Causal Reasoner? A Comprehensive Evaluation [aclanthology.org]
- Large Language Models for Data Annotation: A Survey [arxiv.org]
- Just Ask for Calibration: Strategies for Eliciting Calibrated Confidence Scores from Language Models Fine-Tuned with Human Feedback
- ChatGPT Outperforms Crowd-Workers for Text-Annotation Tasks [arxiv.org]
- LLMs Accelerate Annotation for Medical Information Extraction [proceedings.mlr.press]
- Can Large Language Models Transform Computational Social Science? [arxiv.org]
- LLMAAA: Making Large Language Models as Active Annotators [aclanthology.org]
Other areas to consider could be on how can we better evaluate annotations generated by LLMs, and how to best make use of such annotations along with human annotations, for better training of down-stream models