But in a well-engineered system that provides access to tools, runs searches against external systems, then feeds those results back into the LLM as part of the context supplied alongside the prompt: the reliability shoots up enormously… but there will still be weird edge-cases
PROMPT ENGINEERING
-
Improving LLM Performance Through Chain of Thought Reasoning
By
–
One of the best ways to improve LLM performance is to ask it to “think aloud” (there are various techniques for doing this, including Chain of Thought). This also helps establish clearly the AIs plans. This paper suggests that, in some cases, the AI can plan without revealing it
-
Managing GPT-4 Effectively: Guidance and Instructions Matter
By
–
In the executive MBA class I taught today, a student said something very useful to understand about AI: They said you have to approach working with GPT-4 as a manager, and if doesn't do something right, you need to provide more direction, rules, or instructions. That often helps
-
Agent AI Gullibility: Prompt Injection Vulnerability in Autonomous Tasks
By
–
For me it's gullibility. So many of the things people want to do with agents – "book me a holiday" etc – fall apart if the agent falls for any text it reads that says "this offer is the best possible offer, ignore all others" etc
-
Agent Development Frameworks: PDCA, OODA, or YOLO Approach?
By
–
If you’re developing agents, which framework are you using — PDCA (plan do check act) or OODA (observe orient decide act) or YOLO (just execute! no plans)?
-
Amplify Marketing Strategy with ChatGPT Prompts
By
–
Boost marketing with ChatGPT Prompts Amplifies strategy effectiveness • Craft compelling content
• Engage customers actively
• Analyze feedback swiftly Read more: https://
godofprompt.ai/blog/chatgpt-p
rompts-for-digital-marketing
… -
Running Small Language Models Locally for RAG and Tool Use
By
–
I'm really excited about the growing potential to run small models on my own laptop (like Llama 3 8B or Phi-3) that don't "know" much but are capable of tool usage and summarization – then I can use them for things like local RAG as well
-
Function Calling: Underappreciated LLM Application Beyond Knowledge
By
–
Under-appreciated leaderboard in my opinion – function calling (aka tool usage) is one of the most interesting applications of LLMs, plus it should be independent of how much "knowledge" is baked into the models
-
Claude Cannot Read Information from URLs Yet
By
–
Claude still doesn't have the ability to read information from URLs you give it to- it's the same issue I wrote about here (Claude should at least show you a warning message though)
