Existing solutions to task-oriented dialogue systems follow pipeline designs which introduce architectural complexity and fragility. We propose a novel, holistic, extendable framework based on a single sequence-to-sequence (seq2seq) model which can be optimized with supervised or reinforcement learning. A key contribution is that we design text spans named belief spans to track dialogue believes, allowing task-oriented dialogue systems to be modeled in a seq2seq way. Based on this, we propose a simplistic Two Stage CopyNet instantiation which demonstrates good scalability: significantly reducing model complexity in terms of number of parameters and training time by an order of magnitude. It significantly outperforms state-of-the-art pipeline-based methods on two datasets and retains a satisfactory entity match rate on out-of-vocabulary (OOV) cases where pipeline-designed competitors totally fail.
Knowledge graph reasoning is a critical task in natural language processing. The task becomes more challenging on temporal knowledge graphs, where each fact is associated with a timestamp. Most existing methods focus on reasoning at past timestamps and they are not able to predict facts happening in the future. This paper proposes Recurrent Event Network (RE-NET), a novel autoregressive architecture for predicting future interactions. The occurrence of a fact (event) is modeled as a probability distribution conditioned on temporal sequences of past knowledge graphs. Specifically, our RE-NET employs a recurrent event encoder to encode past facts, and uses a neighborhood aggregator to model the connection of facts at the same timestamp. Future facts can then be inferred in a sequential manner based on the two modules. We evaluate our proposed method via link prediction at future times on five public datasets. Through extensive experiments, we demonstrate the strength of RE-NET, especially on multi-step inference over future timestamps, and achieve state-of-the-art performance on all five datasets 1 .
IntroductionStructural MRI has long been used to characterize local morphological features of the human brain. Coordination patterns of the local morphological features among regions, however, are not well understood. Here, we constructed individual‐level morphological brain networks and systematically examined their topological organization and long‐term test–retest reliability under different analytical schemes of spatial smoothing, brain parcellation, and network type.MethodsThis study included 57 healthy participants and all participants completed two MRI scan sessions. Individual morphological brain networks were constructed by estimating interregional similarity in the distribution of regional gray matter volume in terms of the Kullback–Leibler divergence measure. Graph‐based global and nodal network measures were then calculated, followed by the statistical comparison and intra‐class correlation analysis.ResultsThe morphological brain networks were highly reproducible between sessions with significantly larger similarities for interhemispheric connections linking bilaterally homotopic regions. Further graph‐based analyses revealed that the morphological brain networks exhibited nonrandom topological organization of small‐worldness, high parallel efficiency and modular architecture regardless of the analytical choices of spatial smoothing, brain parcellation and network type. Moreover, several paralimbic and association regions were consistently revealed to be potential hubs. Nonetheless, the three studied factors particularly spatial smoothing significantly affected quantitative characterization of morphological brain networks. Further examination of long‐term reliability revealed that all the examined network topological properties showed fair to excellent reliability irrespective of the analytical strategies, but performing spatial smoothing significantly improved reliability. Interestingly, nodal centralities were positively correlated with their reliabilities, and nodal degree and efficiency outperformed nodal betweenness with respect to reliability.ConclusionsOur findings support single‐subject morphological network analysis as a meaningful and reliable method to characterize structural organization of the human brain; this method thus opens a new avenue toward understanding the substrate of intersubject variability in behavior and function and establishing morphological network biomarkers in brain disorders.
Hate speech classifiers trained on imbalanced datasets struggle to determine if group identifiers like "gay" or "black" are used in offensive or prejudiced ways. Such biases manifest in false positives when these identifiers are present, due to models' inability to learn the contexts which constitute a hateful usage of identifiers. We extract post-hoc explanations from fine-tuned BERT classifiers to detect bias towards identity terms. Then, we propose a novel regularization technique based on these explanations that encourages models to learn from the context of group identifiers in addition to the identifiers themselves. Our approach improved over baselines in limiting false positives on out-of-domain data while maintaining or improving in-domain performance. † * Authors contributed equally † Code is available here "[F]or many Africans, the most threatening kind of ethnic hatred is black against black." -New York Times
Modeling dynamically-evolving, multi-relational graph data has received a surge of interests with the rapid growth of heterogeneous event data. However, predicting future events on such data requires global structure inference over time and the ability to integrate temporal and structural information, which are not yet well understood. We present Recurrent Event Network (RE-NET), a novel autoregressive architecture for modeling temporal sequences of multi-relational graphs (e.g., temporal knowledge graph), which can perform sequential, global structure inference over future time stamps to predict new events. RE-NET employs a recurrent event encoder to model the temporally conditioned joint probability distribution for the event sequences, and equips the event encoder with a neighborhood aggregator for modeling the concurrent events within a time window associated with each entity. We apply teacher forcing for model training over historical data, and infer graph sequences over future time stamps by sampling from the learned joint distribution in a sequential manner. We evaluate the proposed method via temporal link prediction on five public datasets. Extensive experiments 1 demonstrate the strength of RE-NET, especially on multi-step inference over future time stamps.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.