April 2026
Intermediate
395 pages
13h 22m
English
In many cases, using an LLM, even with a simple and poorly thought-out user query, still leads to surprisingly good results. This approach is called direct prompting. But sometimes direct prompting is not enough to achieve the required outcome. But you don’t have to throw in the towel yet because multiple ways are available to tune the system.
Figure 3.16 shows some approaches, from simple to difficult. But the effort might be worth it because with increasing difficulty an increased performance might result.
Figure 3.16 Model Improvement Approaches
The first natural candidate when you’re not satisfied with a model ...
Read now
Unlock full access