Having Your LLM With a Side of Library Science
summary
Claim: inserting the library‑science Big Six steps into prompts makes LLM outputs structured and forces the model to break answers into defined stages, which the presenter calls Blueprinting and follows up with reflective analysis. Demonstrations use ChatGPT, Anthropic Claude and Gemini to generate a Hundred Years War explanation and a neighborhood history and to reparse prior answers. A custom GPT with Big Six context is built and deployed; results look more organized and may cut hallucinations under the Ariadne hypothesis but the effect is not yet proven.
AI-written summary · about
chapters (9)
- 0:00Introduction and welcome
- 1:12Speaker introduction and setup
- 3:30LLMs in workflow and analysis mode
- 5:15Big Six information literacy model
- 8:10Blueprinting technique explained
- 9:56Demo of Blueprinting on different models
- 10:36Reflective analysis technique
- 11:18The Ariadne hypothesis
- 13:47Conclusion and call to action
related talks
47mGive Your LLMs a Left Brain
46mBeyond Basic Prompting: Supercharging Open Source LLMs with LMQL's Structured Generation
35mBuilding LLM-based Agents
15mLLMs for Data-Driven Decision Making in Product Development
01h 15m10x Development: LLMs For the working Programmer
19mUnlocking the Full Potential of LLMs: Advanced Prompt Engineering Techniques
16mLLMs are not the product
37mExpanding the Capabilities of LLMs Through Community Collaboration