How should the advancement of large language models affect the practice of science?

Large language models (LLMs) are being increasingly incorporated into scientific workflows. However, we have yet to fully grasp the implications of this integration. How should the advancement of large language models affect the practice of science? For this opinion piece, we have invited four diverse groups of scientists to reflect on this query, sharing their perspectives and engaging in debate.

Impact of an osteopathic peer recovery coaching model on treatment outcomes in high-risk men entering residential treatment for substance use disorders.

Context: The United States has witnessed a disproportionate rise in substance use disorders (SUD) and co-occurring mental health disorders, paired with housing instability, especially among racially minoritized communities. Traditional in-patient residential treatment programs for SUD have proven inconsistent in their effectiveness in preventing relapse and maintaining attrition among these patient populations. There is evidence showing that peer recovery programs led by individuals who have lived experience with SUD can increase social support and foster intrinsic motivation within participants to bolster their recovery.

Data and its (dis)contents: A survey of dataset development and use in machine learning research.

In this work, we survey a breadth of literature that has revealed the limitations of predominant practices for dataset collection and use in the field of machine learning. We cover studies that critically review the design and development of datasets with a focus on negative societal impacts and poor outcomes for system performance. We also cover approaches to filtering and augmenting data and modeling techniques aimed at mitigating the impact of bias in datasets.