MoPE: Mixture of Prefix Experts for Zero-Shot Dialogue State Tracking

Tianwen Tang, Tong Zhu, Haodong Liu, Yin Bai, Jia Cheng +1 more
4/12/2024
cs.CL

No official code found. Best implementation: ttw1018/MoPE-DST (19 stars, last commit 2025-01-25).

Abstract

Zero-shot dialogue state tracking (DST) transfers knowledge to unseen domains, reducing the cost of annotating new datasets. Previous zero-shot DST models mainly suffer from domain transferring and partial prediction problems. To address these challenges, we propose Mixture of Prefix Experts (MoPE) to establish connections between similar slots in different domains, which strengthens the model transfer performance in unseen domains. Empirical results demonstrate that MoPE-DST achieves the joint goal accuracy of 57.13% on MultiWOZ2.1 and 55.40% on SGD.

View on arXivView PDF

Code Implementations(1)

ttw1018/MoPE-DSTCommunity100%

The code for "MoPE: Mixture of Prefix Experts for Zero-Shot Dialogue State Tracking"

192Shell, PythonMar 23, 20241 years ago

Similar Papers

Diverse and Effective Synthetic Data Generation for Adaptable Zero-Shot Dialogue State Tracking

86% match2024

A Zero-Shot Open-Vocabulary Pipeline for Dialogue Understanding

85% match1 repo32024

Schema Augmentation for Zero-Shot Domain Adaptation in Dialogue State Tracking

83% match2024

Zero-Shot Cross-Domain Dialogue State Tracking via Dual Low-Rank Adaptation

83% match1 repo152024

Plan, Generate and Complicate: Improving Low-resource Dialogue State Tracking via Easy-to-Difficult Zero-shot Data Augmentation

83% match2024

Beyond Single-User Dialogue: Assessing Multi-User Dialogue State Tracking Capabilities of Large Language Models

82% match2025

Cite this paper

@article{tang2024mope,
  title  = {MoPE: Mixture of Prefix Experts for Zero-Shot Dialogue State Tracking},
  author = {Tianwen Tang and Tong Zhu and Haodong Liu and Yin Bai and Jia Cheng and Wenliang Chen},
  year   = {2024},
  eprint = {2404.08559},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2404.08559v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2404.08559.svg)](https://codeofpaper.com/paper/2404.08559)

Discussion