MINDS: A Cross-cultural Dialogue Corpus for Social Norm Classification and Adherence Detection

2025-11-13

AI Summary

Prior work lacks high-quality datasets supporting dynamic modeling of social norms in cross-cultural multi-turn dialogues, and struggles with fine-grained identification of norm categories and compliance states. Method: We introduce MINDS—the first bilingual, multi-turn social norm dialogue dataset—comprising 31 English–Chinese and English–Spanish dialogues, each turn annotated for norm type and adherence. We propose Norm-RAG, a novel framework integrating semantic chunk-based retrieval with agent-style reasoning to jointly model communicative intent, role relations, and linguistic cues in cross-cultural multi-turn settings. Annotation employs multi-annotator consensus and context-aware inference. Contribution/Results: Experiments demonstrate significant improvements in multilingual norm detection accuracy and cross-cultural generalization. MINDS and Norm-RAG establish a new paradigm and foundational resource for culturally adaptive social intelligence dialogue systems.

Technology Category

• Natural Language Processing: Machine Translation, Multilinguality, Cross-Lingual NLP
• Cognitive Modeling & Cognitive Systems: Social Cognition And Interaction
• Multiagent Systems: Agent Communication

Application Category

• Search and Retrieval-Augmented AI: Multilingual and cross-lingual Web search
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Web Mining and Content Analysis: Mining multimedia, multimodal, multilingual, cross-lingual Web data

Abstract

Social norms are implicit, culturally grounded expectations that guide interpersonal communication. Unlike factual commonsense, norm reasoning is subjective, context-dependent, and varies across cultures, posing challenges for computational models. Prior works provide valuable normative annotations but mostly target isolated utterances or synthetic dialogues, limiting their ability to capture the fluid, multi-turn nature of real-world conversations. In this work, we present Norm-RAG, a retrieval-augmented, agentic framework for nuanced social norm inference in multi-turn dialogues. Norm-RAG models utterance-level attributes including communicative intent, speaker roles, interpersonal framing, and linguistic cues and grounds them in structured normative documentation retrieved via a novel Semantic Chunking approach. This enables interpretable and context-aware reasoning about norm adherence and violation across multilingual dialogues. We further introduce MINDS (Multilingual Interactions with Norm-Driven Speech), a bilingual dataset comprising 31 multi-turn Mandarin-English and Spanish-English conversations. Each turn is annotated for norm category and adherence status using multi-annotator consensus, reflecting cross-cultural and realistic norm expression. Our experiments show that Norm-RAG improves norm detection and generalization, demonstrates improved performance for culturally adaptive and socially intelligent dialogue systems.

Problem

• Detecting social norm adherence and violations in multilingual dialogues
• Addressing cultural subjectivity and context-dependency in norm reasoning
• Improving cross-cultural social intelligence in dialogue systems

Innovation

• Retrieval-augmented framework for social norm inference
• Semantic Chunking approach for normative documentation retrieval
• Models utterance attributes for cross-cultural norm reasoning