Evaluating General-Purpose AI with Psychometrics

Natural language processing (NLPs) and Large language models (LLM), such as ChatGPT, represent transformative advancements in artificial intelligence (AI). Their implementation into the medical field has a broad potential, and this review discusses the current trends and prospects of NLPs and LLMs in spine surgery, assessing their potential benefits, applications, and limitations. The methodology involved a comprehensive narrative review of existing English literature related to the use of NLPs and LLMs in spine surgery. We searched the databases PubMed, EMBASE, Web of Science and Scopus from inception until 16th June 2025 using keywords evolving around LLM, natural language processing and spine surgery. Original studies, clinical reports, and case series were included, while abstracts or unpublished studies were excluded. From 221 initial records, 37 studies were included: 18 evaluated LLMs and 19 evaluated NLP-based tools. LLMs were commonly used for clinical decision-making (n = 8), patient counseling (n = 7), classification (n = 2), and in research (n = 1). NLPs were applied in classification tasks (n = 12), clinical decision-making (n = 3), patient counseling (n = 1), postoperative opioid monitoring (n = 2), and research registry development (n = 1). ChatGPT-4 achieved up to 92% accuracy in clinical recommendations, outperforming GPT-3.5 in multiple tasks. Comparative analyses have found that newer versions of LLMs, such as ChatGPT-4, outperform previous versions, evident by greater accuracy and to a lesser extent of artificial hallucination. However, limitations persist, including overconfident outputs, adherence gaps to clinical guidelines, and inconsistent patient readability. While this review suggests that NLPs and LLMs can have a significant impact on spine practice, it is important to keep their limitations in mind and implement them with caution. To maximize the benefits of these models in spine surgery, future research should focus on improving model sensitivity and specificity, promoting multi-disciplinary collaborations, and addressing ethical considerations regarding the use of language models in medical practice, including the inherent issue of hallucination of these models.

BackgroundFeature selection is a critical step for translating advances afforded by systems-scale molecular profiling into actionable clinical insights. While data-driven methods are commonly utilized for selecting candidate genes, knowledge-driven methods must contend with the challenge of efficiently sifting through extensive volumes of biomedical information. This work aimed to assess the utility of large language models (LLMs) for knowledge-driven gene prioritization and selection.MethodsIn this proof of concept, we focused on 11 blood transcriptional modules associated with an Erythroid cells signature. We evaluated four leading LLMs across multiple tasks. Next, we established a workflow leveraging LLMs. The steps consisted of: (1) Selecting one of the 11 modules; (2) Identifying functional convergences among constituent genes using the LLMs; (3) Scoring candidate genes across six criteria capturing the gene’s biological and clinical relevance; (4) Prioritizing candidate genes and summarizing justifications; (5) Fact-checking justifications and identifying supporting references; (6) Selecting a top candidate gene based on validated scoring justifications; and (7) Factoring in transcriptome profiling data to finalize the selection of the top candidate gene.ResultsOf the four LLMs evaluated, OpenAI's GPT-4 and Anthropic's Claude demonstrated the best performance and were chosen for the implementation of the candidate gene prioritization and selection workflow. This workflow was run in parallel for each of the 11 erythroid cell modules by participants in a data mining workshop. Module M9.2 served as an illustrative use case. The 30 candidate genes forming this module were assessed, and the top five scoring genes were identified as BCL2L1, ALAS2, SLC4A1, CA1, and FECH. Researchers carefully fact-checked the summarized scoring justifications, after which the LLMs were prompted to select a top candidate based on this information. GPT-4 initially chose BCL2L1, while Claude selected ALAS2. When transcriptional profiling data from three reference datasets were provided for additional context, GPT-4 revised its initial choice to ALAS2, whereas Claude reaffirmed its original selection for this module.ConclusionsTaken together, our findings highlight the ability of LLMs to prioritize candidate genes with minimal human intervention. This suggests the potential of this technology to boost productivity, especially for tasks that require leveraging extensive biomedical knowledge.

How well will LLMs perform for graph layout tasks?

Recently, the emergence of ChatGPT, an artificial intelligence chatbot developed by OpenAI, has attracted significant attention due to its exceptional language comprehension and content generation capabilities, highlighting the immense potential of large language models (LLMs). LLMs have become a burgeoning hotspot across many fields, including health care. Within health care, LLMs may be classified into LLMs for the biomedical domain and LLMs for the clinical domain based on the corpora used for pre‐training. In the last 3 years, these domain‐specific LLMs have demonstrated exceptional performance on multiple natural language processing tasks, surpassing the performance of general LLMs as well. This not only emphasizes the significance of developing dedicated LLMs for the specific domains, but also raises expectations for their applications in health care. We believe that LLMs may be used widely in preconsultation, diagnosis, and management, with appropriate development and supervision. Additionally, LLMs hold tremendous promise in assisting with medical education, medical writing and other related applications. Likewise, health care systems must recognize and address the challenges posed by LLMs.

Sifting through vast textual data and summarizing key information from electronic health records (EHR) imposes a substantial burden on how clinicians allocate their time. Although large language models (LLMs) have shown immense promise in natural language processing (NLP) tasks, their efficacy on a diverse range of clinical summarization tasks has not yet been rigorously demonstrated. In this work, we apply domain adaptation methods to eight LLMs, spanning six datasets and four distinct clinical summarization tasks: radiology reports, patient questions, progress notes, and doctor-patient dialogue. Our thorough quantitative assessment reveals trade-offs between models and adaptation methods in addition to instances where recent advances in LLMs may not improve results. Further, in a clinical reader study with ten physicians, we show that summaries from our best-adapted LLMs are preferable to human summaries in terms of completeness and correctness. Our ensuing qualitative analysis highlights challenges faced by both LLMs and human experts. Lastly, we correlate traditional quantitative NLP metrics with reader study scores to enhance our understanding of how these metrics align with physician preferences. Our research marks the first evidence of LLMs outperforming human experts in clinical text summarization across multiple tasks. This implies that integrating LLMs into clinical workflows could alleviate documentation burden, empowering clinicians to focus more on personalized patient care and the inherently human aspects of medicine.

Structured representations (SRs), pivotal in pre-LLM NLP, play a contentious role today, with studies showing that they can degrade task performance. The prevailing hypothesis suggests that Large Language Models (LLMs) are unfamiliar with traditional formalisms like Abstract Meaning Representation (AMR). We argue that this view is incomplete, as LLMs are extensively trained on structured data, particularly programming code. To test this, we introduce two new prompt frameworks and evaluate three representation formats (AMR vs. RDF vs. Python code) across multiple LLMs and tasks. Our findings indicate that the choice of representation is of high importance. Across multiple models and tasks, we show that Python code and RDF outperform AMR up to 20% in classification tasks. The effectiveness of any SR is also conditioned on the LLM's baseline capability, the prompting method, and the quality of the representation itself. Although SRs can substantially boost performance for models with weaker baselines, they offer diminishing returns and can harm performance for models that are already highly capable, confirming a "sweet spot" for their application. Our work demonstrates that the utility of SRs in the LLM era depends on their alignment with the models' training data.

This paper assesses the potential for the large language models (LLMs) GPT-4 and GPT-3.5 to aid in deriving insight from education feedback surveys. Exploration of LLM use cases in education has focused on teaching and learning, with less exploration of capabilities in education feedback analysis. Survey analysis in education involves goals such as finding gaps in curricula or evaluating teachers, often requiring time-consuming manual processing of textual responses. LLMs have the potential to provide a flexible means of achieving these goals without specialized machine learning models or fine-tuning. We demonstrate a versatile approach to such goals by treating them as sequences of natural language processing (NLP) tasks including classification (multi-label, multi-class, and binary), extraction, thematic analysis, and sentiment analysis, each performed by LLM. We apply these workflows to a real-world dataset of 2500 end-of-course survey comments from biomedical science courses, and evaluate a zero-shot approach (i.e., requiring no examples or labeled training data) across all tasks, reflecting education settings, where labeled data is often scarce. By applying effective prompting practices, we achieve human-level performance on multiple tasks with GPT-4, enabling workflows necessary to achieve typical goals. We also show the potential of inspecting LLMs’ chain-of-thought (CoT) reasoning for providing insight that may foster confidence in practice. Moreover, this study features development of a versatile set of classification categories, suitable for various course types (online, hybrid, or in-person) and amenable to customization. Our results suggest that LLMs can be used to derive a range of insights from survey text.

Analyzing vast textual data and summarizing key information from electronic health records imposes a substantial burden on how clinicians allocate their time. Although large language models (LLMs) have shown promise in natural language processing (NLP) tasks, their effectiveness on a diverse range of clinical summarization tasks remains unproven. Here we applied adaptation methods to eight LLMs, spanning four distinct clinical summarization tasks: radiology reports, patient questions, progress notes and doctor-patient dialogue. Quantitative assessments with syntactic, semantic and conceptual NLP metrics reveal trade-offs between models and adaptation methods. A clinical reader study with 10 physicians evaluated summary completeness, correctness and conciseness; in most cases, summaries from our best-adapted LLMs were deemed either equivalent (45%) or superior (36%) compared with summaries from medical experts. The ensuing safety analysis highlights challenges faced by both LLMs and medical experts, as we connect errors to potential medical harm and categorize types of fabricated information. Our research provides evidence of LLMs outperforming medical experts in clinical text summarization across multiple tasks. This suggests that integrating LLMs into clinical workflows could alleviate documentation burden, allowing clinicians to focus more on patient care.

Thematic Analysis (TA) is a powerful tool for human factors, HCI, and UX researchers to gather system usability insights from qualitative data like open-ended survey questions. However, TA is both time consuming and difficult, requiring researchers to review and compare hundreds, thousands, or even millions of pieces of text. Recently, this has driven many to explore using Large Language Models (LLMs) to support such an analysis. However, LLMs have their own processing limitations and usability challenges when implementing them reliably as part of a research process – especially when working with a large corpus of data that exceeds LLM context windows. These challenges are compounded when using locally hosted LLMs, which may be necessary to analyze sensitive and/or proprietary data. However, little human factors research has rigorously examined how various prompt engineering techniques can augment an LLM to overcome these limitations and improve usability. Accordingly, in the present paper, we investigate the impact of several prompt engineering techniques on the quality of LLM-mediated TA. Using a local LLM (Llama 3.1 8b) to ensure data privacy, we developed four LLM variants with progressively complex prompt engineering techniques and used them to extract themes from user feedback regarding the usability of a novel knowledge management system prototype. The LLM variants were as follows:1.A “baseline” variant with no prompt engineering or scalability2.A “naïve batch processing” variant that sequentially analyzed small batches of the user feedback to generate a single list of themes3.An “advanced batch processing” variant that built upon the naïve variant by adding prompt engineering techniques (e.g., chain-of-thought prompting)4.A “cognition-inspired” variant that incorporated advanced prompt engineering techniques and kept a working memory-like log of themes and their frequencyContrary to conventional approaches to studying LLMs, which largely rely upon descriptive statistics (e.g., % improvement), we systematically applied a set of evaluation methods from behavioral science and human factors. We performed three stages of evaluation of the outputs of each LLM variant: we compared the LLM outputs to our team’s original TA, we had human factors professionals (N = 4) rate the quality and usefulness of the outputs, and we compared the Inter-Rater Reliability (IRR) of other human factors professionals (N = 2) attempting to code the original data with the outputs generated by each variant. Results demonstrate that even small, locally deployed LLMs can produce high-quality TA when guided by appropriate prompts. While the “baseline” variant performed surprisingly well for small datasets, we found that the other, scalable methods were dependent upon advanced prompt engineering techniques to be successful. Only our novel "cognition-inspired" approach performed as well as the “baseline” variant in qualitative and quantitative comparisons of ratings and coding IRR. This research provides practical guidance for human factors researchers looking to integrate LLMs into their qualitative analysis workflows, disentangling and uncovering the importance of context window limitations, batch processing strategies, and advanced prompt engineering techniques. The findings suggest that local LLMs can serve as valuable and scalable tools in thematic analysis.

Postsurgical falls have significant patient and societal implications but remain challenging to identify and track. Detecting postsurgical falls is crucial to improve patient care for older adults and reduce healthcare costs. Large language models (LLMs) offer a promising solution for reliable and automated fall detection using unstructured data in clinical notes. We tested several LLM prompting approaches to postsurgical fall detection in two different healthcare systems with three open-source LLMs. The Mixtral-8×7B zero-shot had the best performance at Stanford Health Care (PPV = 0.81, recall = 0.67) and the Veterans Health Administration (PPV = 0.93, recall = 0.94). These results demonstrate that LLMs can detect falls with little to no guidance and lay groundwork for applications of LLMs in fall prediction and prevention across many different settings.

Large language models (LLMs) are rapidly advancing medical artificial intelligence, offering revolutionary changes in health care. These models excel in natural language processing (NLP), enhancing clinical support, diagnosis, treatment, and medical research. Breakthroughs, like GPT-4 and BERT (Bidirectional Encoder Representations from Transformer), demonstrate LLMs’ evolution through improved computing power and data. However, their high hardware requirements are being addressed through technological advancements. LLMs are unique in processing multimodal data, thereby improving emergency, elder care, and digital medical procedures. Challenges include ensuring their empirical reliability, addressing ethical and societal implications, especially data privacy, and mitigating biases while maintaining privacy and accountability. The paper emphasizes the need for human-centric, bias-free LLMs for personalized medicine and advocates for equitable development and access. LLMs hold promise for transformative impacts in health care.

PRISMA principles provide a thorough analysis of current advances in large language models (LLMs) and multimodal transformers for medical applications. As LLMs like GPT-4, BioGPT, Med-PaLM, and hybrid frameworks like COMCARE enter clinical processes, thorough synthesis is essential to increase performance, methodological adaptability, and implementation practicality in many healthcare situations. Their creativity in medical report writing, decision support, and diagnosis is notable, but the literature has not established a cohesive taxonomy that evaluates these models by uniform metrics, domain-specific generalizability, and ethical acceptability. Over 40 studies examined radiology report production, clinical question responding, cognitive assessment, and causal reasoning. After testing vision-language transformer architectures like PEGASUS and ETB MII for automated imaging-based reporting, graph-based reasoning was used to evaluate drug safety and interpretability of knowledge- integrated models like KELLM. As needed, BLEU, ROUGE, F1 score, CIDEr, and qualitative evaluations were used. Domain- adapted and hybrid models improve diagnostic accuracy, task- specific explainability, and clinician workload differently. Model illusion, biases, hostile manipulation, and resource-intensive fine- tuning persist. The report recommends strong benchmarking, public evaluation standards, and ethical frameworks for LLMs in high-stakes medical applications. This study defines LLMs' therapeutic utility and recommends infrastructure, ethics, and technology for safe and successful integration. This effort prepares scalable, interpretable, and equitable medical AI systems.

AI systems, particularly large language models (LLMs), are increasingly being employed in high-stakes decisions that impact both individuals and society at large, often without adequate safeguards to ensure safety, quality, and equity. Yet LLMs hallucinate, lack common sense, and are biased-shortcomings that may reflect LLMs' inherent limitations and thus may not be remedied by more sophisticated architectures, more data, or more human feedback. Relying solely on LLMs for complex, high-stakes decisions is therefore problematic. Here, we present a hybrid collective intelligence system that mitigates these risks by leveraging the complementary strengths of human experience and the vast information processed by LLMs. We apply our method to open-ended medical diagnostics, combining 40,762 differential diagnoses made by physicians with the diagnoses of five state-of-the art LLMs across 2,133 text-based medical case vignettes. We show that hybrid collectives of physicians and LLMs outperform both single physicians and physician collectives, as well as single LLMs and LLM ensembles. This result holds across a range of medical specialties and professional experience and can be attributed to humans' and LLMs' complementary contributions that lead to different kinds of errors. Our approach highlights the potential for collective human and machine intelligence to improve accuracy in complex, open-ended domains like medical diagnostics.

In human speakers’ daily conversations, what we do not say matters. We not only compute the literal semantics but also go beyond and draw inferences from what we could have said but chose not to. How well is this pragmatic reasoning process represented in pre-trained large language models (LLM)? In this study, we attempt to address this question through the lens of manner implicature, a pragmatic inference triggered by a violation of the Grice manner maxim. Manner implicature is a central member of the class of context-sensitive phenomena. The current work investigates to what extent pre-trained LLMs are able to identify and tease apart different shades of meaning in manner implicature. We constructed three metrics to explain LLMs’ behavior, including LLMs-surprisals, embedding vectors’ similarities, and natural language prompting. Results showed no striking evidence that LLMs have explainable representations of meaning. First, the LLMs-surprisal findings suggest that some LLMs showed above chance accuracy in capturing different dimensions of meaning, and they were able to differentiate neutral relations from entailment or implications, but they did not show consistent and robust sensitivities to more nuanced comparisons, such as entailment versus implications and equivalence versus entailment. Second, the similarity findings suggest that the perceived advantage of contextual over static embeddings was minimal, and contextual LLMs did not notably outperform static GloVe embeddings. LLMs and GloVe showed no significant difference, though distinctions between entailment and implication were slightly more observable in LLMs. Third, the prompting findings suggest no further supportive evidence indicating LLM’s competence in fully representing different shades of meaning. Overall, our study suggests that current dominant pre-training paradigms do not seem to lead to significant competence in manner implicature within our models. Our investigation sheds light on the design of datasets and benchmark metrics driven by formal and distributional linguistic theories.

This study explores the impact of robot–LLM (Large Language Model) integration on collaborative creative writing, focusing on how embodiment and AI creativity influence various aspects of creative output. A total of 150 undergraduate students participated in a structured experimental design with five collaboration conditions: Human–Human (HH), Human–Computer with High‐Creativity LLM (HC), Human–Robot with High‐Creativity LLM (HR), Human–Robot with Low‐Creativity LLM (RL) and Human–Computer with Low‐Creativity LLM (CL). Creativity was assessed through expert ratings and computational analysis of originality, imagery, voice and semantic flow. The results revealed that while the Human–Robot (High‐Creativity LLM) condition significantly enhanced originality, Human–Human and Human–LLM (text‐based) collaborations excelled in imagery and voice. The study identified an ‘embodiment paradox’, where robot embodiment amplified creativity in high‐creativity AI conditions, yet human collaboration remained superior in stylistic expression. Mediation analysis revealed that user engagement acted as a mediator, with embodiment compensating for low‐creativity AI and amplifying the creative process with high‐creativity AI. The findings have important implications for the design of collaborative AI systems, highlighting the need for a balanced integration of embodiment and AI creativity to optimize creative outcomes. This research contributes to our understanding of how human–robot–LLM collaborations can expand creative potential in writing, offering insights for future AI applications in educational and creative industries. Practitioner notes What is already known about this topic? Previous studies have explored the impact of AI in creative collaborations, with a focus on text‐based models like LLMs enhancing writing quality. Embodiment in AI systems, such as humanoid robots, has been shown to affect user engagement and emotional responses, influencing creativity. Human collaboration has traditionally been seen as superior in generating stylistic elements like imagery and voice, while AI excels in originality and idea generation. What this paper adds? This research demonstrates that robot–LLM collaboration significantly boosts originality, particularly when high‐creativity AI is used. The study uncovers the ‘embodiment paradox’, where embodied robots enhance creativity in high‐creativity AI conditions but human collaboration remains superior in stylistic expression. The mediation role of user engagement is explored, showing how embodiment can enhance creative outcomes when AI creativity is low and amplify them when AI creativity is high. Implications for practice and/or policy Educators and trainers can utilize embodied AI systems in creative tasks to increase student or participant engagement and foster more original outputs. Training programmes can be structured to leverage the strengths of both human collaboration and AI, tailoring tasks based on AI's creativity levels for optimal outcomes. Policy around the integration of AI in educational and creative settings should encourage balanced AI systems that combine embodiment and creativity for enhanced collaborative work.