Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems (2022)


pdf (full)
bib (full)
Proceedings of the Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems (SereTOD)

pdf bib
Proceedings of the Towards Semi-Supervised and Reinforced Task-Oriented Dialog Systems (SereTOD)
Zhijian Ou | Junlan Feng | Juanzi Li

pdf bib
Oh My Mistake!: Toward Realistic Dialogue State Tracking including Turnback Utterances
Takyoung Kim | Yukyung Lee | Hoonsang Yoon | Pilsung Kang | Junseong Bang | Misuk Kim

The primary purpose of dialogue state tracking(DST), a critical component of an end-toend conversational system, is to build a model that responds well to real-world situations. Although we often change our minds from time to time during ordinary conversations, current benchmark datasets do not adequately reflect such occurrences and instead consist of over-simplified conversations, in which no one changes their mind during a conversation. As the main question inspiring the present study, “Are current benchmark datasets sufficiently diverse to handle casual conversations in which one changes their mind after a certain topic is over?” We found that the answer is “No” because DST models cannot refer to previous user preferences when template-based turnback utterances are injected into the dataset. Even in the the simplest mind-changing (turnback) scenario, the performance of DST models significantly degenerated. However, we found that this performance degeneration can be recovered when the turnback scenarios are explicitly designed in the training set, implying that the problem is not with the DST models but rather with the construction of the benchmark dataset.

pdf bib
A GlobalPointer based Robust Approach for Information Extraction from Dialog Transcripts
Yanbo J. Wang | Sheng Chen | Hengxing Cai | Wei Wei | Kuo Yan | Zhe Sun | Hui Qin | Yuming Li | Xiaochen Cai

With the widespread popularisation of intelligent technology, task-based dialogue systems (TOD) are increasingly being applied to a wide variety of practical scenarios. As the key tasks in dialogue systems, named entity recognition and slot filling play a crucial role in the completeness and accuracy of information extraction. This paper is an evaluation paper for Sere-TOD 2022 Workshop challenge (Track 1 Information extraction from dialog transcripts). We proposed a multi-model fusion approach based on GlobalPointer, combined with some optimisation tricks, finally achieved an entity F1 of 60.73, an entity-slot-value triple F1 of 56, and an average F1 of 58.37, and got the highest score in SereTOD 2022 Workshop challenge

A Token-pair Framework for Information Extraction from Dialog Transcripts in SereTOD Challenge
Chenyue Wang | Xiangxing Kong | Mengzuo Huang | Feng Li | Jian Xing | Weidong Zhang | Wuhe Zou

This paper describes our solution for Sere- TOD Challenge Track 1: Information extraction from dialog transcripts. We propose a token-pair framework to simultaneously identify entity and value mentions and link them into corresponding triples. As entity mentions are usually coreferent, we adopt a baseline model for coreference resolution. We exploit both annotated transcripts and unsupervised dialogs for training. With model ensemble and post-processing strategies, our system significantly outperforms the baseline solution and ranks first in triple f1 and third in entity f1.

Prompt Learning for Domain Adaptation in Task-Oriented Dialogue
Makesh Narsimhan Sreedhar | Christopher Parisien

Conversation designers continue to face significant obstacles when creating productionquality task-oriented dialogue systems. The complexity and cost involved in schema development and data collection is often a major barrier for such designers, limiting their ability to create natural, user-friendly experiences. We frame the classification of user intent as the generation of a canonical form, a lightweight semantic representation using natural language. We show that canonical forms offer a promising alternative to traditional methods for intent classification. By tuning soft prompts for a frozen large language model, we show that canonical forms generalize very well to new, unseen domains in a zero- or few-shot setting. The method is also sample-efficient, reducing the complexity and effort of developing new task-oriented dialogue domains.

Disentangling Confidence Score Distribution for Out-of-Domain Intent Detection with Energy-Based Learning
Yanan Wu | Zhiyuan Zeng | Keqing He | Yutao Mou | Pei Wang | Yuanmeng Yan | Weiran Xu

Detecting Out-of-Domain (OOD) or unknown intents from user queries is essential in a taskoriented dialog system. Traditional softmaxbased confidence scores are susceptible to the overconfidence issue. In this paper, we propose a simple but strong energy-based score function to detect OOD where the energy scores of OOD samples are higher than IND samples. Further, given a small set of labeled OOD samples, we introduce an energy-based margin objective for supervised OOD detection to explicitly distinguish OOD samples from INDs. Comprehensive experiments and analysis prove our method helps disentangle confidence score distributions of IND and OOD data.

Semi-Supervised Knowledge-Grounded Pre-training for Task-Oriented Dialog Systems
Weihao Zeng | Keqing He | Zechen Wang | Dayuan Fu | Guanting Dong | Ruotong Geng | Pei Wang | Jingang Wang | Chaobo Sun | Wei Wu | Weiran Xu

Recent advances in neural approaches greatly improve task-oriented dialogue (TOD) systems which assist users to accomplish their goals. However, such systems rely on costly manually labeled dialogs which are not available in practical scenarios. In this paper, we present our models for Track 2 of the SereTOD 2022 challenge, which is the first challenge of building semisupervised and reinforced TOD systems on a large-scale real-world Chinese TOD dataset MobileCS. We build a knowledge-grounded dialog model to formulate dialog history and local KB as input and predict the system response. And we perform semi-supervised pretraining both on the labeled and unlabeled data. Our system achieves the first place both in the automatic evaluation and human interaction, especially with higher BLEU (+7.64) and Success (+13.6%) than the second place.

CMCC: A Comprehensive and Large-Scale Human-Human Dataset for Dialogue Systems
Yi Huang | Xiaoting Wu | Si Chen | Wei Hu | Qing Zhu | Junlan Feng | Chao Deng | Zhijian Ou | Jiangjiang Zhao

Dialogue modeling problems severely limit the real-world deployment of neural conversational models and building a human-like dialogue agent is an extremely challenging task. Recently, data-driven models become more and more prevalent which need a huge amount of conversation data. In this paper, we release around 100,000 dialogue, which come from real-world dialogue transcripts between real users and customer-service staffs. We call this dataset as CMCC (China Mobile Customer Care) dataset, which differs from existing dialogue datasets in both size and nature significantly. The dataset reflects several characteristics of human-human conversations, e.g., task-driven, care-oriented, and long-term dependency among the context. It also covers various dialogue types including task-oriented, chitchat and conversational recommendation in real-world scenarios. To our knowledge, CMCC is the largest real human-human spoken dialogue dataset and has dozens of times the data scale of others, which shall significantly promote the training and evaluation of dialogue modeling methods. The results of extensive experiments indicate that CMCC is challenging and needs further effort. We hope that this resource will allow for more effective models across various dialogue sub-problems to be built in the future.

State-Aware Adversarial Training for Utterance-Level Dialogue Generation
Yi Huang | Xiaoting Wu | Wei Hu | Junlan Feng | Chao Deng

Dialogue generation is a challenging problem because it not only requires us to model the context in a conversation but also to exploit it to generate a coherent and fluent utterance. This paper, aiming for a specific topic of this field, proposes an adversarial training based framework for utterance-level dialogue generation. Technically, we train an encoder-decoder generator simultaneously with a discriminative classifier that make the utterance approximate to the state-aware inputs. Experiments on MultiWoZ 2.0 and MultiWoZ 2.1 datasets show that our method achieves advanced improvements on both automatic and human evaluations, and on the effectiveness of our framework facing low-resource. We further explore the effect of fine-grained augmentations for downstream dialogue state tracking (DST) tasks. Experimental results demonstrate the high-quality data generated by our proposed framework improves the performance over state-of-the-art models.

Information Extraction and Human-Robot Dialogue towards Real-life Tasks A Baseline Study with the MobileCS Dataset
Hong Liu | Hao Peng | Zhijian Ou | Juanzi Li | Yi Huang | Junlan Feng

Recently, there have merged a class of taskoriented dialogue (TOD) datasets collected through Wizard-of-Oz simulated games. However, the Wizard-of-Oz data are in fact simulated data and thus are fundamentally different from real-life conversations, which are more noisy and casual. Recently, the SereTOD challenge is organized and releases the MobileCS dataset, which consists of real-world dialog transcripts between real users and customerservice staffs from China Mobile. Based on the MobileCS dataset, the SereTOD challenge has two tasks, not only evaluating the construction of the dialogue system itself, but also examining information extraction from dialog transcripts, which is crucial for building the knowledge base for TOD. This paper mainly presents a baseline study of the two tasks with the MobileCS dataset. We introduce how the two baselines are constructed, the problems encountered, and the results. We anticipate that the baselines can facilitate exciting future research to build human-robot dialogue systems for real-life tasks.

A Generative User Simulator with GPT-based Architecture and Goal State Tracking for Reinforced Multi-Domain Dialog Systems
Hong Liu | Yucheng Cai | Zhijian Ou | Yi Huang | Junlan Feng

Building user simulators (USs) for reinforcement learning (RL) of task-oriented dialog systems (DSs) has gained more and more attention, which, however, still faces several fundamental challenges. First, it is unclear whether we can leverage pretrained language models to design, for example, GPT-2 based USs, to catch up and interact with the recently advanced GPT- 2 based DSs. Second, an important ingredient in a US is that the user goal can be effectively incorporated and tracked; but how to flexibly integrate goal state tracking and develop an end-to-end trainable US for multi-domains has remained to be a challenge. In this work, we propose a generative user simulator (GUS) with GPT-2 based architecture and goal state tracking towards addressing the above two challenges. Extensive experiments are conducted on MultiWOZ2.1. Different DSs are trained via RL with GUS, the classic agenda-based user simulator (ABUS) and other ablation simulators respectively, and are compared for crossmodel evaluation, corpus-based evaluation and human evaluation. The GUS achieves superior results in all three evaluation tasks.

Offline-to-Online Co-Evolutional User Simulator and Dialogue System
Dafeng Chi | Yuzheng Zhuang | Yao Mu | Bin Wang | Jianzhu Bao | Yasheng Wang | Yuhan Dong | Xin Jiang | Qun Liu | Jianye Hao

Reinforcement learning (RL) has emerged as a promising approach to fine-tune offline pretrained GPT-2 model in task-oriented dialogue (TOD) systems. In order to obtain human-like online interactions while extending the usage of RL, building pretrained user simulators (US) along with dialogue systems (DS) and facilitating jointly fine-tuning via RL becomes prevalent. However, joint training brings distributional shift problem caused by compounding exposure bias. Existing methods usually iterative update US and DS to ameliorate the ensued non-stationarity problem, which could lead to sub-optimal policy and less sample efficiency. To take a step further for tackling the problem, we introduce an Offline-to-oNline Co-Evolutional (ONCE) framework, which enables bias-aware concurrent joint update for RL-based fine-tuning whilst takes advantages from GPT-2 based end-to-end modeling on US and DS. Extensive experiments demonstrate that ONCE builds high-quality loops of policy learning and dialogues data collection, and achieves state-of-the-art online and offline evaluation results on MultiWOZ2.1 dataset. Opensourced code will be implemented with Mindspore (MS, 2022) and released on our homepage.