“Learning to Generate Novel Scientific Directions With Contextualized Literature-Based Discovery”, Qingyun Wang, Doug Downey, Heng Ji, Tom Hope2023-05-23 (GPT-3 nonfiction, GPT-4 nonfiction, T5 Transformer, RL exploration, science)⁠:

Literature-Based Discovery (LBD) aims to discover new scientific knowledge by mining papers and generating hypotheses. Standard LBD is limited to predicting pairwise relations between discrete concepts (eg. drug-disease links), and ignores critical contexts like experimental settings (eg. a specific patient population where a drug is evaluated) and background motivations (eg. to find drugs without specific side effects).

We address these limitations with a novel formulation of contextualized-LBD (C-LBD): generating scientific hypotheses in natural language, while grounding them in a context that controls the hypothesis search space. We present a modeling framework using retrieval of “inspirations” from past scientific papers.

Our evaluations reveal that GPT-4 tends to generate ideas with overall low technical depth and novelty, while our inspiration prompting approaches partially mitigate this issue. Our work represents a first step toward building language models that generate new ideas derived from scientific literature.