Browse Articles

Large language models are increasingly used to solve math problems, but their ability to handle multi-step reasoning remains uncertain. In this study, students tested whether different prompting styles could improve LLM accuracy on challenging AIME math questions and found that detailed step-by-step solutions did not significantly outperform simpler prompts. These results suggest that improving LLM mathematical reasoning may require deeper model-level advances rather than changes in prompting style alone.

The authors looked at how the focus of someone's attention impacted what information was retained in their working memory.

This study explores the interaction between precocene II and trichocethecene 3-O-acetyltransferase using molecular docking simulations. Computational analysis identified several potential binding sites on the enzyme surface and predicted favorable ligand-protein interactions involving key residues. These findings provide insight into how precocene II may interact with this enzyme and demonstrate the use of computational approaches to explore potential antifungal mechanisms.

Here the authors investigated the relationship between offensive play-calling styles and scoring success within the NFL's red zone by analyzing play-by-play data and expected points metrics. Their findings suggest that a conservative approach to play design and execution is more strongly associated with maximizing efficiency and point-value gains than aggressive strategies.

The author looked at how dance styles can vary by individuals, even between a student and their teacher.

The authors survey adults to assess how childhood adversity may impact adult relationships and ways of giving or receiving affection.

This study explored whether planaria, known for their regenerative abilities, can retain learned memories after regeneration and how stressors like alcohol affect memory.

The authors looked at public knowledge regarding suggested daily limits for sugar intake and then looked at how sugar levels vary in the same drink obtained from different sources and across different days.

The authors used Monte Carlo simulations to assess the impacts of various factors on neonatal seizure risk.

The authors developed and tested machine learning methods to diagnose tuberculosis from pulmonary X-ray scans.