研究黑客社群聊天中语言难懂的原因，发现需结合上下文和外部知识才能准确理解。
Understanding Interpretation Difficulty in Harmful Online Communication: Insights from Cybercrime Communities
- 通过专家标注难懂消息的参考解释，评估人与大模型在不同上下文下的理解能力。
- 人类仅靠局部上下文难以理解，需扩展对话和外部知识；大模型在长上下文下表现更优。
- 提出解释困难的分类框架，建议将内容分析视为证据整合问题。
有害网络交流常包含俚语、隐晦术语、缩写及社区特定表达，使信息难以解读。本文对与网络犯罪相关的Discord聊天内容进行探索性研究，选取部分难懂消息并由专家提供参考解释。在此基础上，评估人类与大语言模型（LLM）在不同上下文条件下的理解表现。结果表明，仅依赖局部上下文时人类理解能力有限，而引入外部知识与扩展对话上下文可显著提升理解效果；对于大模型，局部上下文也有助于提升解释能力，且模型规模越大表现越好。我们进一步开展定性错误分析，提出一个初步的解释困难因素分类体系。研究提示，有害内容分析应将解释视为证据整合过程，而非单一消息分类任务。
原文摘要 · Abstract (English)
Harmful online communication often contains slang, coded terms, abbreviations, and community-specific expressions, which make messages difficult to interpret. This paper presents an exploratory study of interpretation difficulty in Discord chats related to cybercrime. We construct reference interpretations of purposefully selected difficult messages, which were reviewed by an expert. We then use them to evaluate human and large language model (LLM) interpretations under different context conditions. The results show that local context alone is often insufficient for humans, while external knowledge and extended conversational context substantially improve human interpretation. For LLMs, local context also improves interpretation, and the larger model performs better. We further conduct a qualitative error analysis and propose a preliminary classification of factors that make harmful chats difficult to interpret. These findings suggest that harmful-content analysis should treat interpretation as an evidence-integration problem, rather than as message-level classification alone.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。