Human-agent Dialogues Research Articles

Towards Informative and Diverse Dialogue Systems Over Hierarchical Crowd Intelligence Knowledge Graph

Knowledge-enhanced dialogue systems aim at generating factually correct and coherent responses by reasoning over knowledge sources, which is a promising research trend. The truly harmonious human-agent dialogue systems need to conduct engaging conversations from three aspects as humans, namely (1) stating factual contents (e.g., records in Wikipedia), (2) conveying subjective and informative opinions about objects (e.g., user discussions on Twitter), and (3) impressing interlocutors with diverse expression styles (e.g., personalized expression habits). The existing knowledge base is a standardized and unified coding for factual knowledge, which could not portray the other two kinds of knowledge to make responses more informative and expressive diverse. To address this, we present CrowdDialog , a crowd intelligence knowledge-enhanced dialogue system, which takes advantage of “crowd intelligence knowledge” extracted from social media (with rich subjective descriptions and diversified expression styles) to promote the performance of dialogue systems. Firstly, to thoroughly mine and organize the crowd intelligence knowledge underlying large-scale and unstructured online contents, we elaborately design the C rowd I ntelligence K nowledge G raph ( CIKG ) structure, including the domain commonsense subgraph, descriptive subgraph, and expressive subgraph. Secondly, to reasonably integrate heterogeneous crowd intelligence knowledge into responses while ensuring logicality and fluency, we propose the G ated F usion with D ynamic Knowledge- D ependent ( GFDD ) model, which generates responses from the semantic and syntactic perspective with the context-aware knowledge gate and dynamic knowledge decoding. Finally, extensive experiments over both Chinese and English dialogue datasets demonstrate that our approach GFDD outperforms competitive baselines in terms of both automatic evaluation and human judgments. Besides, ablation studies indicate that the proposed CIKG has the potential to promote dialogue systems to generate fluent, informative, and diverse dialogue responses.

Towards alignment strategies in human-agent interactions based on measures of lexical repetitions

Alignment of communicative behaviour is an important feature of Human–Human interaction that directly affects the collaboration and the social connection of conversational partners. With the aim of improving the communicative abilities of a virtual agent, and in particular its strategies related to (lexical) verbal alignment, this article focuses on the alignment of linguistic productions of dialogue participants in task-oriented dialogues. We propose a new framework to quantify both the lexical alignment and the self-repetition behaviours of dialogue participants from dyadic dialogue transcripts. The framework involves easily computable measures based on repetition of lexical patterns automatically extracted via a sequential pattern mining approach. These measures allow the characterisation of the nature of these processes by addressing various informative aspects such as their variety, complexity, and strength. This framework is implemented in the freely available and open-source software dialign. Using these measures, we present a contrastive study between Human–Human and Human–Agent dialogues on various corpora that reveals major differences in the lexical alignment and self-repetition behaviours. Lastly, we address the challenge of integrating lexical alignment capabilities in artificial agents. To this end, we describe guidelines and we discuss the integration of the proposed framework in a real-time dialogue system.

Co-constructing Grounded Symbols—Feedback and Incremental Adaptation in Human–Agent Dialogue

Grounding in dialogue concerns the question of how the gap between the individual symbol systems of interlocutors can be bridged so that mutual understanding is possible. This problem is highly relevant to human–agent interaction where mis- or non-understanding is common. We argue that humans minimise this gap by collaboratively and iteratively creating a shared conceptualisation that serves as a basis for negotiating symbol meaning. We then present a computational model that enables an artificial conversational agent to estimate the user’s mental state (in terms of contact, perception, understanding, acceptance, agreement and based upon his or her feedback signals) and use this information to incrementally adapt its ongoing communicative actions to the user’s needs. These basic abilities are important to reduce friction in the iterative coordination process of co-constructing grounded symbols in dialogue.

The Application of a Cognitive Model as a Human Factors Decision Support System for Design Teams

The push in the Navy to design and build new ships that can be effectively manned with dramatically fewer sailors raises key Human Factors issues that must be addressed in the system design process. To promote the understanding of these issues and their solutions throughout the system design team, an effort is currently underway to develop a software agent, called the Executive Advisor (EA). This agent will alert and advise systems engineers and the rest of the engineering team regarding human factors issues and analyses appropriate to the current stage of the system design process. The EA is being built with a cognitive modeling tool called iGEN™, in order to (a) enable context-sensitive reasoning about relevant human factors issues, and (b) facilitate human-agent dialogue. The capabilities and structure of the agent will be described.