In this paper, we introduce the Interpretable Cross-Examination Technique (ICE-T), a novel approach that leverages structured multi-prompt techniques with Large Language Models (LLMs) to improve classification performance over zero-shot and few-shot methods. In domains where interpretability is crucial, such as medicine and law, standard models often fall short due to their "black-box" nature. ICE-T addresses these limitations by using a series of generated prompts that allow an LLM to approach the problem from multiple directions. The responses from the LLM are then converted into numerical feature vectors and processed by a traditional classifier. This method not only maintains high interpretability but also allows for smaller, less capable models to achieve or exceed the performance of larger, more advanced models under zero-shot conditions. We demonstrate the effectiveness of ICE-T across a diverse set of data sources, including medical records and legal documents, consistently surpassing the zero-shot baseline in terms of classification metrics such as F1 scores. Our results indicate that ICE-T can be used for improving both the performance and transparency of AI applications in complex decision-making environments.

The need for health providers to share information is a pressing need in our ever more connected world. A patient's health information should seamlessly flow from labs to hospitals to primary care offices. To address this need, in this paper we present the Health E-Match, which focuses on the matching health terms in support of semantic interoperability. Health E-Match determines the semantic similarity between data items, realizing, for instance, that "BHGC (UR)" and "BETA-HCG (QUAL)" both refer to the same pregnancy test, known as "Beta human chorionic gonadotropin, urine qualitative." Our approach is grounded in probabilistic machine learning, and leverages several sophisticated methods for comparing the similarity between medical data items beyond simple edit distance. We present two large scale, real-world experiments to verify that our approach is both accurate and has the ability to eventually be "universal" in that models trained on one set of data translate to strong performance on data from a completely different provider.

The "Journal of Artificial Intelligence Research (JAIR) was one of the first scientific journals distributed over the web. It has now completed over five years of successful publication. Electronic publishing is reshaping the way academic work is disseminated, and JAIR is leading the way toward a future where scientific articles are freely and easily accessible to all. This report describes how the journal has evolved, its "grassroots" philosophy, and prospects for the future.

Ellen L. Walker Linda T. Powell counter.