Search for: All records

The long-standing one-to-many problem of gold standard responses in open-domain dialogue systems presents challenges for automatic evaluation metrics. Though prior works have demonstrated some success by applying powerful Large Language Models (LLMs), existing approaches still struggle with the one-to-many problem, and exhibit subpar performance in domain-specific scenarios. We assume the commonsense reasoning biases within LLMs may hinder their performance in domain-specific evaluations. To address both issues, we propose a novel framework SLIDE (Small and Large Integrated for Dialogue Evaluation), that leverages both a small, specialized model (SLM), and LLMs for the evaluation of open-domain dialogues. Our approach introduces several techniques: (1) Contrastive learning to differentiate between robust and non-robust response embeddings; (2) A novel metric for semantic sensitivity that combines embedding cosine distances with similarity learned through neural networks, and (3) A strategy for incorporating the evaluation results from both the SLM and LLMs. Our empirical results demonstrate that our approach achieves state-of-the-art performance in both the classification and evaluation tasks, and additionally, the SLIDE evaluator exhibits a better correlation with human judgments. 

The MRI-derived brain network serves as a pivotal instrument in elucidating both the structural and functional aspects of the brain, encompassing the ramifications of diseases and developmental processes. However, prevailing methodologies, often focusing on synchronous BOLD signals from functional MRI (fMRI), may not capture directional influences among brain regions and rarely tackle temporal functional dynamics. In this study, we first construct the brain-effective network via the dynamic causal model. Subsequently, we introduce an interpretable graph learning framework termed Spatio-Temporal Embedding ODE (STE-ODE). This framework incorporates specifically designed directed node embedding layers, aiming at capturing the dynamic interplay between structural and effective networks via an ordinary differential equation (ODE) model, which characterizes spatial-temporal brain dynamics. Our framework is validated on several clinical phenotype prediction tasks using two independent publicly available datasets (HCP and OASIS). The experimental results clearly demonstrate the advantages of our model compared to several state-of-the-art methods. 

We interpret the Taylor–Green cellular vortex model in terms of the Kolmogorov length and velocity scales, in order to study the balance between aggregation and breakup of cohesive sediment in fine-scale turbulence. One-way coupled numerical simulations, which capture the effects of cohesive, lubrication and direct contact forces on the flocculation process, reproduce the non-monotonic relationship between the equilibrium floc size and shear rate observed in previous experiments. The one-way coupled results are confirmed by select two-way coupled simulations. Intermediate shear gives rise to the largest flocs, as it promotes preferential concentration of the primary particles without generating sufficiently strong turbulent stresses to break up the emerging aggregates. We find that the optimal intermediate shear rate increases for stronger cohesion and smaller particle-to-fluid density ratios, and we propose a simple model for the equilibrium floc size that agrees well with experimental data reported in the literature.