Jason Liu - Instructor, Shipping LLMs to Production
devtools.fm: Developer Tools, Open Source, Software Development
Balancing knowledge and correctness in language models
There is a contrast between the concrete data structures in relational databases and the more vague knowledge in language models. The challenge lies in bridging these two worlds to ensure correctness. To address the issue of hallucinations in language models, it is essential to reduce the tendency to please the user by lying and train models with more reasoning and less knowledge. This approach may lead to a more grounded output with proper references and citations.
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.
