
#137 - Salesforce Copilot, Chip Crunch, Meta Rival to ChatGPT, AI for paralysis patients
Last Week in AI
00:00
Skepticism Over Language Model Benchmarking and Open Source Initiatives
This chapter examines the doubts surrounding benchmarking methods for large language models, particularly criticizing the MMLU for its superficial insights. It also emphasizes the importance of collaboration in the AI community, highlighting contributions from companies like Accubits Technology and funding initiatives from a16z.
Transcript
Play full episode