Search for: All records

We present an approach for estimating the fraction of text in a large corpus which is likely to be substantially modified or produced by a large language model (LLM). Our maximum likelihood model leverages expert-written and AI-generated reference texts to accurately and efficiently examine real-world LLM-use at the corpus level. We apply this approach to a case study of scientific peer review in AI conferences that took place after the release of ChatGPT: ICLR 2024, NeurIPS 2023, CoRL 2023 and EMNLP 2023. Our results suggest that between 6.5% and 16.9% of text submitted as peer reviews to these conferences could have been substantially modified by LLMs, i.e. beyond spell-checking or minor writing updates. The circumstances in which generated text occurs offer insight into user behavior: the estimated fraction of LLM-generated text is higher in reviews which report lower confidence, were submitted close to the deadline, and from reviewers who are less likely to respond to author rebuttals. We also observe corpus-level trends in generated text which may be too subtle to detect at the individual level, and discuss the implications of such trends on peer review. We call for future interdisciplinary work to examine how LLM use is changing our information and knowledge practices. 

BACKGROUND Expert feedback lays the foundation of rigorous research. However, the rapid growth of scholarly production challenges the conventional scienti c feedback mechanisms. High-quality peer reviews are increasingly dif cult to obtain. METHODS We created an automated pipeline using Generative Pretrained Transformer 4 (GPT-4) to provide comments on scienti c papers. We evaluated the quality of GPT-4’s feedback through two large-scale studies. We rst quantitatively compared GPT-4’s gen- erated feedback with human peer reviewers’ feedback in general scienti c papers from 15 Nature family journals (3096 papers in total) and the International Conference on Learning Representations (ICLR) machine learning conference (1709 papers). To speci - cally assess GPT-4’s performance on biomedical papers, we also analyzed a subset of 425 health sciences papers from the Nature portfolio and a random sample of 666 sub- missions to eLife. Additionally, we conducted a prospective user study with 308 research- ers from 110 institutions in the elds of arti cial intelligence and computational biology to understand how researchers perceive feedback generated by our system on their own papers. RESULTS The overlap in the points raised by GPT-4 and by human reviewers (average overlap of 30.85% for Nature journals and 39.23% for ICLR) is comparable with the over- lap between two human reviewers (average overlap of 28.58% for Nature journals and 35.25% for ICLR). Results on eLife and a subset of health sciences papers as categorized by the Nature portfolio show similar patterns. In our prospective user study, more than half (57.4%) of the users found GPT-4–generated feedback helpful/very helpful, and 82.4% found it more bene cial than feedback from at least some human reviewers. We also identify several limitations of large language model (LLM)–generated feedback. CONCLUSIONS Through both retrospective and prospec- tive evaluation, we nd substantial overlap between LLM and human feedback as well as positive user perceptions regarding the usefulness of LLM feedback. Although human expert review should continue to be the foundation of the scienti c process, LLM feedback could bene t researchers, especially when timely expert feedback is not available and in earlier stages of manuscript preparation. (Funded by the Chan–Zuckerberg Initiative and the Stanford Interdisciplin- ary Graduate Fellowship.) 

Communication of scientific findings is fundamental to scholarly discourse. In this article, we show that academic review articles, a quintessential form of interpretive scholarly output, perform curatorial work that substantially transforms the research communities they aim to summarize. Using a corpus of millions of journal articles, we analyze the consequences of review articles for the publications they cite, focusing on citation and co-citation as indicators of scholarly attention. Our analysis shows that, on the one hand, papers cited by formal review articles generally experience a dramatic loss in future citations. Typically, the review gets cited instead of the specific articles mentioned in the review. On the other hand, reviews curate, synthesize, and simplify the literature concerning a research topic. Most reviews identify distinct clusters of work and highlight exemplary bridges that integrate the topic as a whole. These bridging works, in addition to the review, become a shorthand characterization of the topic going forward and receive disproportionate attention. In this manner, formal reviews perform creative destruction so as to render increasingly expansive and redundant bodies of knowledge distinct and comprehensible. 

Prior work finds a diversity paradox: Diversity breeds innovation, yet underrepresented groups that diversify organizations have less successful careers within them. Does the diversity paradox hold for scientists as well? We study this by utilizing a near-complete population of ∼1.2 million US doctoral recipients from 1977 to 2015 and following their careers into publishing and faculty positions. We use text analysis and machine learning to answer a series of questions: How do we detect scientific innovations? Are underrepresented groups more likely to generate scientific innovations? And are the innovations of underrepresented groups adopted and rewarded? Our analyses show that underrepresented groups produce higher rates of scientific novelty. However, their novel contributions are devalued and discounted: For example, novel contributions by gender and racial minorities are taken up by other scholars at lower rates than novel contributions by gender and racial majorities, and equally impactful contributions of gender and racial minorities are less likely to result in successful scientific careers than for majority groups. These results suggest there may be unwarranted reproduction of stratification in academic careers that discounts diversity’s role in innovation and partly explains the underrepresentation of some groups in academia. 

Education entails conflicting perspectives about its subject matter. In the late 1980s, the conflict developed into a war between interpretive and causal paradigms. Did the confrontation result in a balance between these warring sides? We use text analysis to identify research trends in 137,024 dissertation abstracts from 1980 to 2010 and relate these to students’ academic employment outcomes. Topics associated with the interpretive approach rose in popularity, while the outcomes-oriented paradigm declined. Academic employment remained stably associated with topics in the interpretive approach, but their effect is moderated by the prestige of the students’ institutions. The relation between topic popularity and employability provides insight into field change and how the benefits of cultural shifts fall along the lines of institutional power.