Search for: All records

The instruction hierarchy, which establishes a priority order from system messages to user messages, conversation history, and tool outputs, is essential for ensuring consistent and safe behavior in language models (LMs). Despite its importance, this topic receives limited attention, and there is a lack of comprehensive benchmarks for evaluating models’ ability to follow the instruction hierarchy. We bridge this gap by introducing IHEval, a novel benchmark comprising 3,538 examples across nine tasks, covering cases where instructions in different priorities either align or conflict. Our evaluation of popular LMs highlights their struggle to recognize instruction priorities. All evaluated models experience a sharp performance decline when facing conflicting instructions, compared to their original instruction-following performance. Moreover, the most competitive open-source model only achieves 48% accuracy in resolving such conflicts. Our results underscore the need for targeted optimization in the future development of LMs. 

Multi-modal data are prevalent in many scientific fields. In this study, we consider the parameter estimation and variable selection for a multi-response regression using block-missing multi-modal data. Our method allows the dimensions of both the responses and the predictors to be large, and the responses to be incomplete and correlated, a common practical problem in high-dimensional settings. Our proposed method uses two steps to make a prediction from a multi-response linear regression model with block-missing multi-modal predictors. In the first step, without imputing missing data, we use all available data to estimate the covariance matrix of the predictors and the cross-covariance matrix between the predictors and the responses. In the second step, we use these matrices and a penalized method to simultaneously estimate the precision matrix of the response vector, given the predictors, and the sparse regression parameter matrix. Lastly, we demonstrate the effectiveness of the proposed method using theoretical studies, simulated examples, and an analysis of a multi-modal imaging data set from the Alzheimer’s Disease Neuroimaging Initiative. 

Abstract We have successfully synthesized ultrathin nanowires of pure Pt, Pt₉₉Ni₁, Pt₉Ni₁, and Pt₇Ni₃using a modified room‐temperature soft‐template method. Analysis of both methanol oxidation reaction (MOR) and ethanol oxidation reaction (EOR) results found that the Pt₇Ni₃samples yielded the best performance with specific activities of 0.36 and 0.34 mA/cm²respectively. Additionally, formic acid oxidation reaction (FAOR) tests noted that both Pt and PtNi nanowires oxidize small organic molecules (SOMs) via an indirect pathway. CO oxidation data suggests little measurable performance without any pre‐reduction treatment; however, after annealing in H₂, we detected significantly improved CO₂formation for both Pt₉Ni₁and Pt₇Ni₃motifs. These observations highlight the importance of pre‐treating these nanowires under a reducing atmosphere to enhance their performance for CO oxidation. To explain these findings, we collected extended x‐ray adsorption fine structure (EXAFS) spectroscopy data, consistent with the presence of partial alloying with a tendency for Pt and Ni to segregate, thereby implying the formation of a Pt‐rich shell coupled with a Ni‐rich core. We also observed that the degree of alloying within the nanowires increased after annealing in a reducing atmosphere, a finding deduced through analysis of the coordination numbers and calculations of Cowley's short range order parameters.