Jan 15, 2022 · In this paper, we focus on how to utilize the language understanding and generation ability of pre-trained language models for DST. We design a ...

A prompt learning framework for few-shot DST, which consists of two main components: value-based prompt and inverse prompt mechanism, that indicates that ...

ABSTRACT · References

Mar 16, 2022 · In this work, we propose an in-context learning (ICL) framework for zero-shot and few-shot learning DST, where a large pre-trained language ...

In this work, we propose an in-context (IC) learning framework for zero-shot and few-shot learning dialogue state tracking (DST), where a large pretrained ...

A novel dual prompt learning framework is designed to help PLMs in understanding the essence of DST with few labels and utilize the generation ability of PLMs ...

In this paper, we focus on how to learn a DST model efficiently with limited labeled data. We design a prompt learning framework for few-shot DST, which ...

First, we reformulate DST as a text-to-SQL task, in- cluding a tabular description of the ontology in the prompt. This is a better match to the knowledge-.

The task is to track the slot values associated with a user request up to the current turn (dialogue state). In few-shot settings, given a test turn, IC-DST ...

In this paper, we propose a compressive sensing based framework for robust visual tracking. As a key part of the tracking framework, a new multi-task sparse ...