OpenAI's Erdos result renews focus on two papers testing reasoning in large language models with chain-of-thought prompting on GSM8K benchmarks
Engineer credits papers as debate notes prior use of techniques.
——0——
Engineer credits papers as debate notes prior use of techniques.
Many users praised early CoT papers as foundational to today's LLM reasoning advances that aged well after earlier dismissal, while one viewed the OpenAI Erdos spotlight as mere flag planting.
6 comments with sentiment.
