Skip to main content

Derek Chiang

Derek Chiang is VP and Head of Computational Biology (Oncology) at Bayer Pharmaceuticals.

Nov 11, 2026

20:19
I wonder, so along those lines, if you could point to an instance where a model of your work was like confidently wrong, or there was an instance where it felt like this idea of accuracy and evidence was something you had to make a judgment call on or how you might handle that type of a situation.
20:44
I can give an example that's a year old, which I used as a warning about blind trust to large language models.
20:53
The hard part is that this example is now a year out of date and the models develop so quickly that it's no longer a concern.
21:02
And I'll explain both parts of my answer.
21:05
I refer to the complexity of a question.
21:09
I actually really like Bloom's taxonomy of cognitive capabilities, which is a mouthful.
21:16
But it was established by educators in the 1950s after the GI Bill on what should higher education teach.

6 MINS LATER

26:57
So what are some of the ideas around model correctness and the evaluation of model correctness that you and your colleagues are talking about?

We value your privacy

We use cookies to understand how you use our platform and to improve your experience. Click “Accept All” to consent, or “Decline non-essential” to opt out of non-essential cookies. Read our Privacy Policy.