rm abstract addt outline

Author: kdillmcfarland

content/speakers/speaker_info_2025/regular/yan_liu/index.md

@@ -14,81 +14,7 @@ Beyond technical implementation the session will cover essential prompt engineer
 
 By the end of this session participants will be equipped with practical tools best practices and insights into the evolving role of LLMs in R development. Attendees will leave with a clear understanding of how to apply these techniques to enhance the quality and efficiency of their own analytical work. 
 
-Outline 
 
-1. Introduction (2 min) 
-
-Why LLMs Matter for R Developers 
-
-Automating text analysis and content generation 
-
-Efficiency gains in qualitative research and evaluation 
-
-What This Talk Covers 
-
-Three practical use cases using {ellmer} 
-
-Hands-on examples with key takeaways 
-
- 
-
-2. Quick Overview of LLMs & {ellmer} (2 min) 
-
-What are LLMs? (Brief high-level) 
-
-Local (ollama) vs Cloud models (OpenAI) 
-
-Why Use {ellmer}? 
-
-Bridges R workflows with LLMs 
-
-Comparison with other R text analysis tools 
-
- 
-
-3. Practical Applications: Three Use Cases (9 min total 3 min each) 
-
-Application 1: Image-to-Text Generation (3 min) 
-
-Problem: Extracting structured text from images for research or automation 
-
-Example Solution: Using {ellmer} provide a written description of a ggplot 
-
-Key Considerations: Accuracy of the summary prompt design 
-
- 
-
-Application 2: Text Classification (3 min) 
-
-Problem: Automating categorization of qualitative responses 
-
-Example Solution: Using {ellmer} to classify text based on themes 
-
-Key Considerations: Model reliability handling ambiguous responses 
-
- 
-
-Application 3: Summarization (3 min) 
-
-Problem: Condensing lengthy documents into key insights 
-
-Example Solution: {ellmer} prompt for summarizing reports/articles 
-
-Key Considerations: Balancing brevity vs. detail avoiding bias in summaries 
-
- 
-
-4. Closing Thoughts & Takeaways (2 min) 
-
-Key Learnings 
-
-LLMs can streamline text-based workflows in R 
-
-{ellmer} provides a simple way to integrate LLMs 
-
-Best practices for prompt design and model selection 
-
-Where to Learn More (Resources package documentation) 
 <br><br>
 
 <table>