Choice Fusion as Knowledge for Zero-Shot Dialogue State Tracking

This is the PyTorch implementation of the following paper accepted by ICASSP 2023 in Rhodes Island, Greek:

Title: Choice Fusion as Knowledge for Zero-Shot Dialogue State Tracking

Authors: Ruolin Su, Jingfeng Yang, Ting-Wei Wu, Biing-Hwang Juang.

Usage

Install Dependency

conda creat -n py38 python=3.8
conda activate py38
pip install -r requirements.txt

Download and Create the Question-Answering Dataset

Download RACE dataset and put it under qa_data folder. Download other QA datasets:

./download_data.sh

Combine and pre-process the QA datasets:

python create_qa_data.py

Download and Create the MultiWOZ2.1 Dataset

python create_data_mwoz.py

Train

1. Train our model with appreciated-choice selection:

./run_qa_pretrain_t5.sh pretrain

2. Train our model with choice fusion mechanism including appreciated-choice selection and context-knowledge fusion:

./run_qa_pretrain_t5.sh pretrain_fusion

--percentage The percentage of combined QA data for training. --max_seq_length The max length of the input tokens. --num_train_epochs The num of training epochs. --overwrite_cache Whether or not use the cached training dataset. The number of CUDA_VISIBLE_DEVICES, --per_device_train_batch_size and --gradient_accumulation_steps Multiply to get the total batch size. --neg_num --neg_context_ratio Negative sampling rate to encourage generating none values proactively. link

run_qa_pretrain_t5.sh: (1) pretrain. --percentage --evaluation_strategy --eval_steps --save_strategy --save_steps (2) predict.

Evaluation

./run_qa_pretrain_t5.sh predict

--history_turn Previous turns used as the dialogue context for test. --per_device_eval_batch_size The batch size for test. --test_type dst for evaluating on the test set of MultiWOZ, or qa for evaluating on the QA dev set. --overwrite_cache Whether or not use the cached DST test dataset. To generate DST slot-values with the trained context-knowledge fusion model, run ./run_qa_pretrain_t5.sh predict_fusion.