For the practical deployment of the Role-Playing Agents with Large Language Models, prior work has mainly focused on suppressing out-of-character behavior relative to explicitly provided character information such as personas. However, as shown in psychological research, out-of-character behavior can arise not only from contradictions with explicit information but also from mismatches with character images implicitly inferred by users. Modeling user-inference-dependent out-of-character behavior is a challenging task, because it is unclear what kinds of inferences users make from existing information. To tackle this problem, we construct a dataset of 480 dialogue scenarios and collect 2,152 descriptions from users familiar with those titles. Using cases that annotators judged inappropriate despite the absence of contradiction with existing settings, we derive a 12-category typology, organized into general, title-specific, and individual-specific groups. We also conduct an exploratory evaluation of category-guided feedback, finding no consistent overall advantage over standard self-refinement but possible benefits for some characters. Our shareable code and dataset are at https://github.com/CyberAgentAILab/User-Inference-OOC.

Challenges in multimodal task-oriented dialogue between humans and systems, particularly those involving audio and visual interactions, have not been sufficiently explored or shared, forcing researchers to define improvement directions individually without a clearly shared roadmap. To address these challenges, we organized a competition for multimodal task-oriented dialogue systems and constructed a large competition-based dataset of 1,865 minutes of Japanese task-oriented dialogues. This dataset includes audio and visual interactions between diverse systems and human participants. After analyzing system behaviors identified as problematic by the human participants in questionnaire surveys and notable methods employed by the participating teams, we identified key challenges in multimodal task-oriented dialogue systems and discussed potential directions for overcoming these challenges.

Generating and evaluating character-like utterances automatically is essential for applications ranging from character simulation to creative-writing support. Existing approaches primarily focus on basic aspects of character‐likeness, such as script-fidelity knowledge and conversational ability. However, achieving a higher level of character‐likeness in utterance generation and evaluation requires consideration of the character’s identity, which deeply reflects the character’s inner self. To bridge this gap, we identified a set of identity-centric character-likeness elements. First, we listed 27 elements covering various aspects of identity, drawing on psychology and identity theory. Then, to clarify the features of each element, we collected utterances annotated with these elements from a commercial smartphone game and analyzed them based on user evaluations regarding character-likeness and charm. Our analysis reveals part of element-wise effects on character‐likeness and charm. These findings enable developers to design practical and interpretable element-feature-aware generation methods and evaluation metrics for character-like utterances.

In human-human conversation, interpersonal consideration for the interlocutor is essential, and similar expectations are increasingly placed on dialogue systems. This study examines the behavior of dialogue systems in a specific interpersonal scenario where a user vents frustrations and seeks emotional support from a long-time friend represented by a dialogue system. We conducted a human evaluation and qualitative analysis of 15 dialogue systems under this setting. These systems implemented diverse strategies, such as structuring dialogue into distinct phases, modeling interpersonal relationships, and incorporating cognitive behavioral therapy techniques. Our analysis reveals that these approaches contributed to improved perceived empathy, coherence, and appropriateness, highlighting the importance of design choices in socially sensitive dialogue.

Chat-oriented dialogue systems that deliver tangible benefits, such as sharing news or frailty prevention for seniors, require proactive acquisition of specific user information via chats on user-favored topics. This study proposes the Proactive Information Acquisition (PIA) task to support the development of these systems. In this task, a system needs to acquire a user’s answers to predefined questions without making the user feel abrupt while engaging in a chat on a predefined topic. We created and analyzed a dataset of 650 PIA chats, identifying key challenges and effective strategies for recent LLMs. Our system, designed from these insights, surpassed the performance of LLMs prompted solely with task instructions. Finally, we demonstrate that automatic evaluation of this task is reasonably accurate, suggesting its potential as a framework to efficiently develop techniques for systems dealing with complex dialogue goals, extending beyond the scope of PIA alone. Our dataset is available at: https://github.com/CyberAgentAILab/PIA