In this study, we investigated how the complexity of a query can influence tool-use hallucinations in large language models that were tasked with setting an appointment. Using explainable AI (XAI) as a tool to test the hypothesis, we analyzed trends in hallucination rates by the type of query and saw results that when complexity and ambiguity of the query increased (such as asking for multiple days or unclear time ranges), the model was more likely to make errors due to misunderstanding time-related data from tool outputs. This shows that making large language models (LLMs) understand what seems logical to a human is still difficult, especially in complex tasks, and that knowing what causes these errors can help build more trustworthy AI systems.
The authors use machine learning to analyze electroencephalogram data and identify slowing patterns that can indicate undetected disorders like epilepsy or dementia
Although there has been great progress in the field of Natural language processing (NLP) over the last few years, particularly with the development of attention-based models, less research has contributed towards modeling keystroke log data. State of the art methods handle textual data directly and while this has produced excellent results, the time complexity and resource usage are quite high for such methods. Additionally, these methods fail to incorporate the actual writing process when assessing text and instead solely focus on the content. Therefore, we proposed a framework for modeling textual data using keystroke-based features. Such methods pay attention to how a document or response was written, rather than the final text that was produced. These features are vastly different from the kind of features extracted from raw text but reveal information that is otherwise hidden. We hypothesized that pairing efficient machine learning techniques with keystroke log information should produce results comparable to transformer techniques, models which pay more or less attention to the different components of a text sequence in a far quicker time. Transformer-based methods dominate the field of NLP currently due to the strong understanding they display of natural language. We showed that models trained on keystroke log data are capable of effectively evaluating the quality of writing and do it in a significantly shorter amount of time compared to traditional methods. This is significant as it provides a necessary fast and cheap alternative to increasingly larger and slower LLMs.
Cardiac autonomic and structural changes may occur in temporal lobe epilepsy patients and contribute to the risk of sudden unexpected death in epilepsy patients. Choi and colleagues reviewed clinical charts to obtain patients’ lifetime seizure count, antiepileptic drug use, and history of heart disease, followed by transthoracic echocardiogram to calculate left ventricle dimensions, ejection fraction, and left ventricle mass. By comparing epilepsy patients to control subjects, they found that epilepsy patients had thinner left ventricle walls and smaller ejection fraction, but with no significant difference in left ventricle mass.
The authors looked at genes that were differentially expressed in patients who had and had not had febrile seizures to determine what differences in gene expression existed.
Image credit: Meister, Horvath, and Brown de Colstoun
This manuscript investigates the urban heat island (UHI) effect by utilizing two satellite datasets: Landsat (high spatial resolution, lower temporal resolution) and MODIS (lower spatial resolution, high temporal resolution). The authors hypothesized that Landsat would provide better spatial detail, while MODIS would better capture temporal variations. Their analysis in the Washington D.C.–Baltimore region supports these hypotheses, demonstrating that Landsat offers finer spatial details, whereas MODIS provides more consistent seasonal patterns and better detects heatwave frequencies.
Control of voluntary and involuntary movements is one of the most important aspects of human neurological function, but the mechanisms of motor control are not completely understood. In this study, the authors use transcranial magnetic stimulation (TMS) to stimulate a portion of the motor cortex while subjects performed either discrete (e.g. throwing) or rhythmic (e.g. walking) movements. By recording electrical activity in the muscles during this process, the authors showed that motor evoked potentials (MEPs) measured in the muscles during TMS stimulation are larger in amplitude for discrete movements than for rhythmic movements. Interestingly, they also found that MEPs during transitions between rhythmic and discrete movements were nearly identical and larger in amplitude than those recorded during either rhythmic or discrete movements. This research provides important insights into the mechanisms of neurological control of movement and will serve as the foundation for future studies to learn more about temporal variability in neural activity during different movement types.
In this study, the authors use bioinformatic approaches to characterize the mirror neurons, which are active when performing and seeing certain actions. They also investigated whether mirror neuron impairment was connected to neural degenerative diseases and psychiatric disorders.
Music is an important part of the day for many of us, but what makes some combinations of sound more pleasant to the ear than others? In this article the authors investigate the role of some characteristics of sound on the perception of a pleasant vs harsh musical note.
