Task-oriented dialog (TOD) systems often need to formulate knowledge base (KB) queries corresponding the user intent and use query results generate system responses. Existing approaches require datasets explicitly annotate these KB -- annotations can be time consuming, expensive. In response, we define novel problems of predicting training agent, without explicit annotation. For prediction, pro...

Abstract Direct decoding for task-oriented dialogue is known to suffer from the explaining-away effect, manifested in models that prefer short and generic responses. Here we argue use of Bayes’ theorem factorize task into two models, distribution context given response, prior response itself. This approach, an instantiation noisy channel model, both mitigates effect allows principled incorporat...

Research on the effectiveness of group support systems (GSS) technology has begun to explore the “task/technology” fit hypothesis, which suggests an interaction between task type and communication mode. Some limited research employing generic tasks and ad hoc groups has found that GSS-mediated groups perform about the same as their face-to-face counterparts on creative tasks but significantly w...

Communication systems to date primarily aim at reliably communicating bit sequences. Such an approach provides efficient engineering designs that are agnostic the meanings of messages or goal message exchange aims achieve. Next generation systems, however, can be potentially enriched by folding semantics and goals communication into their design. Further, these made cognizant context in which t...

Abstract Data hallucination generates additional training examples for novel classes to alleviate the data scarcity problem in few‐shot learning (FSL). Existing hallucination‐based FSL methods normally train a general embedding model first by applying information extracted from base that have abundant data. In those methods, hallucinators are then built upon trained generate classes. However, t...