Tag: LLM
-
Language model hallucination can still be accurate
I wanted to test if Claude AI could read and summarise an article when only given a URL. According to the response from the model, Claude can’t visit links. However, its summary of the article at the URL is spot on. Like, really good. So either Claude is lying and can visit links, or it’s…
-
Article: Towards Expert-Level Medical Question Answering with Large Language Models
Singhal, K., et al. (2023). Towards Expert-Level Medical Question Answering with Large Language Models (arXiv:2305.09617). arXiv. From the abstract: We performed detailed human evaluations on long-form questions along multiple axes relevant to clinical applications. In pairwise comparative ranking of 1066 consumer medical questions, physicians preferred Med-PaLM 2 answers to those produced by physicians on eight…
-
Generative AI in higher education: Paradigm shifts in assessment – Cape Peninsula University of Technology symposium
accountability, ai, artificial intelligence, assessment, assessment concerns, assessment design, assessment paradigm, assessment task, assessment validity, faculty development, higher education, large language model, learning inference, LLM, paradigm shift, pedagogy, standard assessment paradigm, universal anything machineIn this presentation at the Cape Peninsula University of Technology, I examines the impact of large language models (LLMs) on assessment practices in professional education. I critique the standard assessment paradigm and suggest that AI could reshape assessment methods. The presentation also briefly covers a faculty development framework and broader implications of AI in learning.
-
Bing allows you to modulate the amount of ‘hallucination’ in your response
Last week I wrote about LLM hallucinations, and how this isn’t the problem that everyone thinks it is. “I expect that soon we’ll see language models with features that allow us to modulate the output in some way. We may want to dial up creativity or serendipity, in which case we’ll see less overlap with…
-
Universal ‘anything’ machines – North-West University symposium
In this presentation I describe the concept of “universal ‘anything’ machines,” which leverage large language models (LLMs) capable of producing human-like text across various domains. These LLMs enable the creation of customisable, context-aware characters with expertise in multiple disciplines, accessible via natural language. The talk also addresses the implications for higher education, discussing the potential…
-
Tutoring through large language models doesn’t need to be perfect
We know the gold standard for learning is one-to-one tutoring, where a more experienced expert providing focused guidance to a novice. This kind of apprenticeship model worked well for thousands of years, but doesn’t scale to meet the requirements of modern institutions. However, large language models and generative AI can provide reasonably high-level expertise in…