Program-of-Thought Prompting Outperforms Chain-of-Thought by 15% (2022)
https://arxiv.org/abs/2211.12588
#HackerNews #ProgramOfThought #Prompting #ChainOfThought #AIResearch #MachineLearning #2022Study
#Tag
Program-of-Thought Prompting Outperforms Chain-of-Thought by 15% (2022)
https://arxiv.org/abs/2211.12588
#HackerNews #ProgramOfThought #Prompting #ChainOfThought #AIResearch #MachineLearning #2022Study
Stop using prompt checklists. They failed 20 years ago, and they're failing now.
When the internet arrived, critical reading checklists (like CRAAP) became obsolete because humans are unlikely to apply static rules to dynamic content.
Same story with AI. Sequential optimization (conversation) crushes single-shot prompts.
The future isn't better prompts. It's better dialogue.
https://wiobyrne.com/weve-been-thinking-about-prompts-all-wrong/
Honda: 2 years of ml vs 1 month of prompting - heres what we learned
https://www.levs.fyi/blog/2-years-of-ml-vs-1-month-of-prompting/
#HackerNews #Honda #ml #prompting #machinelearning #AI #insights
A space for Bonfire maintainers and contributors to communicate