Stop Chunking Like It's 2022 — Yuval Belfer, AI21 Labs
Chunking remains relevant despite trends favoring agentic search methods.
“I will claim that if we have to kill something if something has to be dead then it's probably retrieval tuning.”
Yuval Belfer of AI21 Labs argues that while many believe chunking is outdated due to the rise of agentic search, it still has a crucial role in handling large data sets effectively. The emphasis on chunking suggests that AI development is not simply about adopting new trends, but also understanding the foundational principles that continue to apply.