Discourse & Dialogue
Robots-Dont-Cry: Understanding Falsely Anthropomorphic Utterances in Dialog Systems
Dialog systems are often designed or trained to output human-like responses. However, some responses may be impossible for a machine to truthfully say (e.g. "that movie made me cry"). Highly anthropomorphic responses might make users uncomfortable or implicitly deceive them into thinking they are interacting with a human. We collect human ratings on the feasibility of approximately 900 two-turn dialogs sampled from 9 diverse data sources. Ratings are for two hypothetical machine embodiments: a futuristic humanoid robot and a digital assistant. We find that for some data-sources commonly used to train dialog systems, 20-30% of utterances are not viewed as possible for a machine. Rating is marginally affected by machine embodiment. We explore qualitative and quantitative reasons for these ratings. Finally, we build classifiers and explore how modeling configuration might affect output permissibly, and discuss implications for building less falsely anthropomorphic dialog systems.
Robustifying Sentiment Classification by Maximally Exploiting Few Counterfactuals
De Raedt, Maarten, Godin, Fréderic, Develder, Chris, Demeester, Thomas
For text classification tasks, finetuned language models perform remarkably well. Yet, they tend to rely on spurious patterns in training data, thus limiting their performance on out-of-distribution (OOD) test data. Among recent models aiming to avoid this spurious pattern problem, adding extra counterfactual samples to the training data has proven to be very effective. Yet, counterfactual data generation is costly since it relies on human annotation. Thus, we propose a novel solution that only requires annotation of a small fraction (e.g., 1%) of the original training data, and uses automatic generation of extra counterfactuals in an encoding vector space. We demonstrate the effectiveness of our approach in sentiment classification, using IMDb data for training and other sets for OOD tests (i.e., Amazon, SemEval and Yelp). We achieve noticeable accuracy improvements by adding only 1% manual counterfactuals: +3% compared to adding +100% in-distribution training samples, +1.3% compared to alternate counterfactual approaches.
A Dialogue Robot System to Improve Credibility in Sightseeing Spot Recommendations
Yoshimaru, Naoki, Masuda, Tomohiro, Hong, Hyejin, Tanaka, Yusei, Okuma, Motoharu, Matsumoto, Nagihiro, Kusu, Kazuma, Iio, Takamasa, Hatano, Kenji
Various studies have been conducted on human-supporting robot systems. These systems have been put to practical use over the years and are now seen in our daily lives. In particular, robots communicating smoothly with people are expected to play an active role in customer service and guidance. In this case, it is essential to determine whether the customer is satisfied with the dialog robot or not. However, it is not easy to satisfy all of the customer's requests due to the diversity of the customer's speech. In this study, we developed a dialog mechanism that prevents dialog breakdowns and keeps the customer satisfied by providing multiple scenarios for the robot to take control of the dialog. We tested it in a travel destination recommendation task at a travel agency.
Back to the Future: Bidirectional Information Decoupling Network for Multi-turn Dialogue Modeling
Li, Yiyang, Zhao, Hai, Zhang, Zhuosheng
Multi-turn dialogue modeling as a challenging branch of natural language understanding (NLU), aims to build representations for machines to understand human dialogues, which provides a solid foundation for multiple downstream tasks. Recent studies of dialogue modeling commonly employ pre-trained language models (PrLMs) to encode the dialogue history as successive tokens, which is insufficient in capturing the temporal characteristics of dialogues. Therefore, we propose Bidirectional Information Decoupling Network (BiDeN) as a universal dialogue encoder, which explicitly incorporates both the past and future contexts and can be generalized to a wide range of dialogue-related tasks. Experimental results on datasets of different downstream tasks demonstrate the universality and effectiveness of our BiDeN.
Sentiment-Aware Word and Sentence Level Pre-training for Sentiment Analysis
Fan, Shuai, Lin, Chen, Li, Haonan, Lin, Zhenghao, Su, Jinsong, Zhang, Hang, Gong, Yeyun, Guo, Jian, Duan, Nan
Most existing pre-trained language representation models (PLMs) are sub-optimal in sentiment analysis tasks, as they capture the sentiment information from word-level while under-considering sentence-level information. In this paper, we propose SentiWSP, a novel Sentiment-aware pre-trained language model with combined Word-level and Sentence-level Pre-training tasks. The word level pre-training task detects replaced sentiment words, via a generator-discriminator framework, to enhance the PLM's knowledge about sentiment words. The sentence level pre-training task further strengthens the discriminator via a contrastive learning framework, with similar sentences as negative samples, to encode sentiments in a sentence. Extensive experimental results show that SentiWSP achieves new state-of-the-art performance on various sentence-level and aspect-level sentiment classification benchmarks. We have made our code and model publicly available at https://github.com/XMUDM/SentiWSP.
Spoken Dialogue Strategy Focusing on Asymmetric Communication with Android Robots
Kawakubo, Daisuke, Ishii, Hitoshi, Okazawa, Riku, Nishizawa, Shunta, Hatakeyama, Haruki, Sugiyama, Hiroaki, Shuzo, Masaki, Maeda, Eisaku
Humans are easily conscious of small differences in an android robot's (AR's) behaviors and utterances, resulting in treating the AR as not-human, while ARs treat us as humans. Thus, there exists asymmetric communication between ARs and humans. In our system at Dialogue Robot Competition 2022, this asymmetry was a considerable research target in our dialogue strategy. For example, tricky phrases such as questions related to personal matters and forceful requests for agreement were experimentally used in AR's utterances. We assumed that these AR phrases would have a reasonable chance of success, although humans would likely hesitate to use the phrases. Additionally, during a five-minute dialogue, our AR's character, such as its voice tones and sentence expressions, changed from mechanical to human-like type in order to pretend to tailor to customers. The characteristics of the AR developed by our team, DSML-TDU, are introduced in this paper.
Information Extraction and Human-Robot Dialogue towards Real-life Tasks: A Baseline Study with the MobileCS Dataset
Liu, Hong, Peng, Hao, Ou, Zhijian, Li, Juanzi, Huang, Yi, Feng, Junlan
Recently, there have merged a class of task-oriented 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 customer-service 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 Data-Driven Investigation of Noise-Adaptive Utterance Generation with Linguistic Modification
Chingacham, Anupama, Demberg, Vera, Klakow, Dietrich
In noisy environments, speech can be hard to understand for humans. Spoken dialog systems can help to enhance the intelligibility of their output, either by modifying the speech synthesis (e.g., imitate Lombard speech) or by optimizing the language generation. We here focus on the second type of approach, by which an intended message is realized with words that are more intelligible in a specific noisy environment. By conducting a speech perception experiment, we created a dataset of 900 paraphrases in babble noise, perceived by native English speakers with normal hearing. We find that careful selection of paraphrases can improve intelligibility by 33% at SNR -5 dB. Our analysis of the data shows that the intelligibility differences between paraphrases are mainly driven by noise-robust acoustic cues. Furthermore, we propose an intelligibility-aware paraphrase ranking model, which outperforms baseline models with a relative improvement of 31.37% at SNR -5 dB.
A Generative User Simulator with GPT-based Architecture and Goal State Tracking for Reinforced Multi-Domain Dialog Systems
Liu, Hong, Cai, Yucheng, Ou, Zhijian, Huang, Yi, Feng, Junlan
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 cross-model evaluation, corpus-based evaluation and human evaluation. The GUS achieves superior results in all three evaluation tasks.
Team Flow at DRC2022: Pipeline System for Travel Destination Recommendation Task in Spoken Dialogue
Hirai, Ryu, Ohashi, Atsumoto, Guo, Ao, Shiroma, Hideki, Zhou, Xulin, Tone, Yukihiko, Iizuka, Shinya, Higashinaka, Ryuichiro
To improve the interactive capabilities of a dialogue system, e.g., to adapt to different customers, the Dialogue Robot Competition (DRC2022) was held. As one of the teams, we built a dialogue system with a pipeline structure containing four modules. The natural language understanding (NLU) and natural language generation (NLG) modules were GPT-2 based models, and the dialogue state tracking (DST) and policy modules were designed on the basis of hand-crafted rules. After the preliminary round of the competition, we found that the low variation in training examples for the NLU and failed recommendation due to the policy used were probably the main reasons for the limited performance of the system.