Baseline Text Testing

[bishalu]

Prompt for ChatGPT:

We are developing a plug-in to simplify complex texts. I need to find 5 different paragraphs to simplify as a baseline test set to test our simplification algorithm. Please provide 5 paragraphs from landmark papers, books or other literature with complicated text for the following topics. Please include citations:

1. Philosophy by Kant
2. Machine Learning related to Generative AI
3. Physics related to relativity
4. Literature from Shakespere with old English such as "thou"
5. Neuroscience related to memory

Then include a simplified version of every text. Every original and simplified section of text should be around 200 words.

Result:

Philosophy by Kant

Original: "That all our knowledge begins with experience there can be no doubt. For how is it possible that the faculty of cognition should be awakened into exercise otherwise than by means of objects which affect our senses, and partly of themselves produce representations, partly rouse our powers of understanding into activity, to compare to connect, or to separate these, and so to convert the raw material of our sensuous impressions into a knowledge of objects, which is called experience? In respect of time, therefore, no knowledge of ours is antecedent to experience, but begins with it." (Kant, Immanuel. Critique of Pure Reason. Cambridge University Press, 1998)
Simplified: Without a doubt, our understanding of things starts with experience. This is because our senses need to be stimulated by objects to start processing information. These objects create images in our mind and activate our ability to understand, allowing us to relate, link, or differentiate these images. We then transform these raw sensory impressions into knowledge about things - we call this process experience. So, in terms of time, we don't know anything before we experience it - all knowledge starts from experience.
Machine Learning related to Generative AI

Original: "Generative models are a subclass of unsupervised learning that generates new sample/data by learning and mimicking the underlying data distribution. These models are widely used for learning the true data distribution of a training set so as to generate new data with some variations. Generative models contrast with discriminative models, where the latter classify input examples." (Goodfellow, Ian, et al. Generative adversarial nets. Advances in neural information processing systems, 2014)
Simplified: Generative models are a type of machine learning where the goal is to create new data. They do this by studying and copying the structure of the data they're trained on. These models are often used to learn the pattern of a training set and then generate new data that's similar but not exactly the same. This is different from discriminative models, which focus on classifying input data.
Physics related to relativity

Original: "The basic tenet of the general theory of relativity, namely, that gravitation is merely a manifestation of the curvature of space-time by matter, can be precisely stated in mathematical form. The curvature of space-time at any event in the gravitational field is determined completely by the matter at that event. Specifically, the curvature is proportional to the energy-momentum tensor." (Einstein, Albert. The Meaning of Relativity. Princeton University Press, 1950)
Simplified: The key idea in the general theory of relativity is that gravity is just a result of the bending of space and time by matter. This idea can be clearly stated with mathematics. How space and time curve at any point in a gravity field is totally decided by the matter at that point. Specifically, the curve is related to the amount of energy and momentum.
Literature from Shakespeare

Original: "But, soft! what light through yonder window breaks? It is the east, and Juliet is the sun. Arise, fair sun, and kill the envious moon, Who is already sick and pale with grief, That thou, her maid, art far more fair than she." (Shakespeare, William. Romeo and Juliet. Oxford University Press, 2000)
Simplified: Wait, what's that light coming from the window over there? It's coming from the east, and Juliet is like the sun. Rise up, beautiful sun, and outshine the jealous moon. The moon is already looking sick and sad because you, Juliet, are much more beautiful than it is.
Neuroscience related to memory

Original: "The human memory system comprises multiple components. The ability to hold information in mind in the short term (working memory) is encoded in persistent neural activity in a prefrontal-parietal network; the ability to form new long-term memories (declarative memory), in regionally specific modifications in synaptic strength, leading to durable changes in the firing patterns of interconnected neural ensembles." (Eichenbaum, Howard. The cognitive neuroscience of memory: an introduction. Oxford University Press, 2011)
Simplified: The human memory system has many parts. Short-term memory, or the capacity to hold information in our mind for a short while (also known as working memory), is maintained by continuous brain activity in a network between the frontal and parietal brain areas. The ability to form new long-term memories (known as declarative memory), involves changes in the strength of connections between neurons in specific brain areas, causing lasting changes in how these groups of neurons fire together.