During task-oriented dialogues (TODs), human users naturally introduce chitchat that is beyond the immediate scope of the task, interfering with the flow of the conversation. To address this issue without the need for expensive manual data creation, we use few-shot prompting with Llama-2-70B to enhance the MultiWOZ dataset with user backstories, a typical example of chitchat interference in TODs. We assess the impact of this addition by testing two models: one trained solely on TODs and another trained on TODs with a preliminary chitchat interaction. Our analysis demonstrates that our enhanced dataset poses a challenge for these systems. Moreover, we demonstrate that our dataset can be effectively used for training purposes, enabling a system to consistently acknowledge the user's backstory while also successfully moving the task forward in the same turn, as confirmed by human evaluation. These findings highlight the benefits of generating novel chitchat-TOD scenarios to test TOD systems more thoroughly and improve their resilience to natural user interferences

As a recent development, task-oriented dialogues (TODs) have been enriched with chitchat in an effort to make dialogues more diverse and engaging. This enhancement is particularly valuable as TODs are often confined to narrow domains, making the mitigation of repetitive and predictable responses a significant challenge. This paper presents a comparative analysis of three chitchat enhancements, aiming to identify the most effective approach in terms of diversity. Additionally, we quantify the divergence between the added chitchat, the original task-oriented language, and chitchat typically found in chitchat datasets, highlighting the top 20 divergent keywords for each comparison. Our findings drive a discussion on future enhancements for augmenting TODs, emphasizing the importance of grounding dialogues beyond the task to achieve more diverse and natural exchanges.

Recent years have seen increasing concerns about the unsafe response generation of large-scale dialogue systems, where agents will learn offensive or biased behaviors from the real-world corpus. Some methods are proposed to address the above issue by detecting and replacing unsafe training examples in a pipeline style. Though effective, they suffer from a high annotation cost and adapt poorly to unseen scenarios as well as adversarial attacks. Besides, the neglect of providing safe responses (e.g. simply replacing with templates) will cause the information-missing problem of dialogues. To address these issues, we propose an unsupervised pseudo-label sampling method, TEMP, that can automatically assign potential safe responses. Specifically, our TEMP method groups responses into several clusters and samples multiple labels with an adaptively sharpened sampling strategy, inspired by the observation that unsafe samples in the clusters are usually few and distribute in the tail. Extensive experiments in chitchat and task-oriented dialogues show that our TEMP outperforms state-of-the-art models with weak supervision signals and obtains comparable results under unsupervised learning settings.

Keyphrase extraction is one of the essential tasks for document understanding in NLP. While the majority of the prior works are dedicated to the formal setting, e.g., books, news or web-blogs, informal texts such as video transcripts are less explored. To address this limitation, in this work we present a novel corpus and method for keyphrase extraction from the transcripts of the videos streamed on the Behance platform. More specifically, in this work, a novel data augmentation is proposed to enrich the model with the background knowledge about the keyphrase extraction task from other domains. Extensive experiments on the proposed dataset dataset show the effectiveness of the introduced method.

Machine-learning based dialogue managers are able to learn complex behaviors in order to complete a task, but it is not straightforward to extend their capabilities to new domains. We investigate different policies' ability to handle uncooperative user behavior, and how well expertise in completing one task (such as restaurant reservations) can be reapplied when learning a new one (e.g. booking a hotel). We introduce the Recurrent Embedding Dialogue Policy (REDP), which embeds system actions and dialogue states in the same vector space. REDP contains a memory component and attention mechanism based on a modified Neural Turing Machine, and significantly outperforms a baseline LSTM classifier on this task. We also show that both our architecture and baseline solve the bAbI dialogue task, achieving 100% test accuracy.

Tom was discussing the film star Tang Wei with a chatbot named XiaoIce, and the bot was excited: "A goddess! She stole my heart … and then went off and married!" Married who? "Haven't you heard?" XiaoIce replied. "Tang Wei is engaged to famous Korean director Kim Tae-yong." XiaoIce is a massive hit on social networks in Asia. Introduced in 2014 by Microsoft Research and Bing in Beijing, it can answer simple questions, like a stripped-down version of Cortana.

Training is the foundation of data-driven smarts. The conversational intelligence of virtual digital assistants--aka chatbots--depends on the extent to which their statistical algorithms have been trained with the most relevant, high-quality data for the task at hand. Without frequent retraining on fresh data, even the most expertly scripted chatbot will behave like a clueless dummy. Fortunately for chatbot developers, training resources are amply available for building and tuning the smarts of your AI-driven digital assistants 24 7. If you're building these bots into your mobile, social, e-commerce, Internet of Things, and other apps, here are the training options you should explore: Of course, composing chatbots is as much of a conversational art--akin to screenwriting or ventriloquism--as it is a data science. To do it well, developers need to ensure that all this technical wizardry is concealed by a seemingly simple, natural, friendly, fun, and useful interface.

We present a general solution towards building task-oriented dialogue systems for online shopping, aiming to assist online customers in completing various purchase-related tasks, such as searching products and answering questions, in a natural language conversation manner. As a pioneering work, we show what & how existing NLP techniques, data resources, and crowdsourcing can be leveraged to build such task-oriented dialogue systems for E-commerce usage. To demonstrate its effectiveness, we integrate our system into a mobile online shopping app. To the best of our knowledge, this is the first time that an AI bot in Chinese is practically used in online shopping scenario with millions of real consumers. Interesting and insightful observations are shown in the experimental part, based on the analysis of human-bot conversation log. Several current challenges are also pointed out as our future directions.

The Tap is similar to Echo, but requires you to press a button to speak to Alexa. SAN FRANCISCO – Amazon is offering 1 million to the university team that builds an artificial intelligence that can keep up its side of the conversation with a human being for 20 minutes. On Thursday Amazon announced the Alexa Prize, a 1 million award for the creation of a conversational artificial intelligence that can talk to people "coherently and engagingly" for a third of an hour. Such a system would be unprecedented "and at least five times more advanced than state-of-the-art conversational AI," said Rohit Prasad, vice president and head scientist of Amazon Alexa. To aid the endeavor, up to ten teams will get a 100,000 stipend from Amazon along with Alexa-enabled devices, free cloud computing and support from Amazon's Alexa team.