paper-with-me

Papers

TaSL: Continual Dialog State Tracking via Task Skill Localization and Consolidation

2024-08-19 · Yujie Feng, Xu Chu, Yongxin Xu, Guangyuan Shi, Bo Liu, Xiao-Ming Wu

A practical dialogue system requires the capacity for ongoing skill acquisition and adaptability to new tasks while preserving prior knowledge. However, current methods for Continual Dialogue State Tracking (DST), a crucial function of dialogue systems, struggle with the catastrophic forgetting issue and knowledge transfer between tasks. We present TaSL, a novel framework for task skill localization and consolidation that enables effective knowledge transfer without relying on memory replay. TaSL uses a novel group-wise technique to pinpoint task-specific and task-shared areas. Additionally, a fine-grained skill consolidation strategy protects task-specific knowledge from being forgotten while updating shared knowledge for bi-directional knowledge transfer. As a result, TaSL strikes a balance between preserving previous knowledge and excelling at new tasks. Comprehensive experiments on various backbones highlight the significant performance improvements of TaSL over existing state-of-the-art methods. The source code is provided for reproducibility.

📄 PDF Abstract BibTeX arXiv:2408.09857

Code (1)

WoodScene/TaSL 공식 구현 pytorch

Tasks

dialog state trackingDialogue State TrackingTransfer Learning

Similar Papers 제목 키워드 기반

Continual Prompt Tuning for Dialog State Tracking

2021-11-16 · ACL ARR November 2021 11 · Anonymous

A desirable dialog system should be able to continually learn new skills without forgetting old ones, and thereby adapt to new domains or tasks in its life cycle. However, continually training a model often leads to a we…

Continual Learningdialog state trackingTransfer Learning

Continual Prompt Tuning for Dialog State Tracking

2022-03-13 · ACL 2022 5 · Qi Zhu, Bing Li, Fei Mi, Xiaoyan Zhu 외

A desirable dialog system should be able to continually learn new skills without forgetting old ones, and thereby adapt to new domains or tasks in its life cycle. However, continually training a model often leads to a we…

Continual Learningdialog state trackingTransfer Learning

Continual Dialogue State Tracking via Example-Guided Question Answering

2023-05-23 · Hyundong Cho, Andrea Madotto, Zhaojiang Lin, Khyathi Raghavi Chandu 외

Dialogue systems are frequently updated to accommodate new services, but naively updating them by continually training with data for new services in diminishing performance on previously learnt services. Motivated by the…

Continual LearningDialogue State TrackingMemorizationNatural Language Understanding+1

Prompt Pool based Class-Incremental Continual Learning for Dialog State Tracking

2023-11-17 · Hong Liu, Yucheng Cai, Yuan Zhou, Zhijian Ou 외

Continual learning is crucial for dialog state tracking (DST) in dialog systems, since requirements from users for new functionalities are often encountered. However, most of existing continual learning methods for DST r…

Continual Learningdialog state tracking

Domain-Lifelong Learning for Dialogue State Tracking via Knowledge Preservation Networks

2021-11-01 · EMNLP 2021 11 · Qingbin Liu, Pengfei Cao, Cao Liu, Jiansong Chen 외

Dialogue state tracking (DST), which estimates user goals given a dialogue context, is an essential component of task-oriented dialogue systems. Conventional DST models are usually trained offline, which requires a fixed…

Dialogue State TrackingDiversityKnowledge DistillationLifelong learning+1