Mikhail Burstev, Founder of Deep Pavlov, talks AI orchestration and LLMs
VUX World
Large Language Models for Intent Classification and Entity Extraction
The large language models are performing worse than a specifically tuned classifier for that specific task. Even if there is all information in the input the model still can produce errors. So it's inherently like because large language models they have trained for so many different tasks. They have like it's inherent inability to focus narrowly on some specific task.
00:00
Transcript
Play full episode
Remember Everything You Learn from Podcasts
Save insights instantly, chat with episodes, and build lasting knowledge - all powered by AI.
