paper-with-me

Papers

Query Enhanced Knowledge-Intensive Conversation via Unsupervised Joint Modeling

2022-12-19 · Mingzhu Cai, Siqi Bao, Xin Tian, Huang He, Fan Wang, Hua Wu

In this paper, we propose an unsupervised query enhanced approach for knowledge-intensive conversations, namely QKConv. There are three modules in QKConv: a query generator, an off-the-shelf knowledge selector, and a response generator. QKConv is optimized through joint training, which produces the response by exploring multiple candidate queries and leveraging corresponding selected knowledge. The joint training solely relies on the dialogue context and target response, getting exempt from extra query annotations or knowledge provenances. To evaluate the effectiveness of the proposed QKConv, we conduct experiments on three representative knowledge-intensive conversation datasets: conversational question-answering, task-oriented dialogue, and knowledge-grounded conversation. Experimental results reveal that QKConv performs better than all unsupervised methods across three datasets and achieves competitive performance compared to supervised methods.

📄 PDF Abstract BibTeX arXiv:2212.09588

Code (3)

PaddlePaddle/Knover/tree/develop/projects/QKConv 공식 구현 paddle
MindCode-4/code-12/tree/main/model-conversion-via mindspore
MindSpore-scientific/code-6/tree/main/multi-adversarial-domain-adaptation mindspore

Tasks

Conversational Question AnsweringQuestion AnsweringRetrieval

Similar Papers 제목 키워드 기반

Conversational Query Rewriting with Self-supervised Learning

2021-02-09 · Hang Liu, Meng Chen, Youzheng Wu, Xiaodong He 외

Context modeling plays a critical role in building multi-turn dialogue systems. Conversational Query Rewriting (CQR) aims to simplify the multi-turn dialogue modeling into a single-turn problem by explicitly rewriting th…

Self-Supervised Learning

KIMAs: A Configurable Knowledge Integrated Multi-Agent System

2025-02-13 · Zitao Li, Fei Wei, Yuexiang Xie, Dawei Gao 외

Knowledge-intensive conversations supported by large language models (LLMs) have become one of the most popular and helpful applications that can assist people in different aspects. Many current knowledge-intensive appli…

ManagementRAGRetrievalRetrieval-augmented Generation

Hint-enhanced In-Context Learning wakes Large Language Models up for knowledge-intensive tasks

2023-11-03 · Yifan Wang, Qingyan Guo, Xinzhe Ni, Chufan Shi 외

In-context learning (ICL) ability has emerged with the increasing scale of large language models (LLMs), enabling them to learn input-label mappings from demonstrations and perform well on downstream tasks. However, unde…

In-Context LearningOpen-Domain Question AnsweringQuestion Answering

Tackling Query-Focused Summarization as A Knowledge-Intensive Task: A Pilot Study

2021-12-14 · Weijia Zhang, Svitlana Vakulenko, Thilina Rajapakse, Yumo Xu 외

Query-focused summarization (QFS) requires generating a summary given a query using a set of relevant documents. However, such relevant documents should be annotated manually and thus are not readily available in realist…

Open-Domain Question AnsweringPassage RetrievalQuery-focused SummarizationQuestion Answering+2

TREC CAsT 2019: The Conversational Assistance Track Overview

2020-03-30 · Jeffrey Dalton, Chenyan Xiong, Jamie Callan

The Conversational Assistance Track (CAsT) is a new track for TREC 2019 to facilitate Conversational Information Seeking (CIS) research and to create a large-scale reusable test collection for conversational search syste…

Conversational SearchLearning-To-RankMachine Reading ComprehensionReading Comprehension+2