[论文解读] Multi-Agent Causal Reasoning for Suicide Ideation Detection Through Online Conversations
论文提出 MACR,一种使用反事实推理和前门因果校正来从在线对话树预测自杀风险的两智能体框架,性能优于基线。
Suicide remains a pressing global public health concern. While social media platforms offer opportunities for early risk detection through online conversation trees, existing approaches face two major limitations: (1) They rely on predefined rules (e.g., quotes or relies) to log conversations that capture only a narrow spectrum of user interactions, and (2) They overlook hidden influences such as user conformity and suicide copycat behavior, which can significantly affect suicidal expression and propagation in online communities. To address these limitations, we propose a Multi-Agent Causal Reasoning (MACR) framework that collaboratively employs a Reasoning Agent to scale user interactions and a Bias-aware Decision-Making Agent to mitigate harmful biases arising from hidden influences. The Reasoning Agent integrates cognitive appraisal theory to generate counterfactual user reactions to posts, thereby scaling user interactions. It analyses these reactions through structured dimensions, i.e., cognitive, emotional, and behavioral patterns, with a dedicated sub-agent responsible for each dimension. The Bias-aware Decision-Making Agent mitigates hidden biases through a front-door adjustment strategy, leveraging the counterfactual user reactions produced by the Reasoning Agent. Through the collaboration of reasoning and bias-aware decision making, the proposed MACR framework not only alleviates hidden biases, but also enriches contextual information of user interactions with counterfactual knowledge. Extensive experiments on real-world conversational datasets demonstrate the effectiveness and robustness of MACR in identifying suicide risk.
研究动机与目标
- 在在线对话中超越仅记日志交互的鲁棒自杀风险检测动机。
- 缓解如用户一致性与模仿行为等不可观测混淆因素。
- 开发一个可扩展的框架,通过反事实推理丰富上下文。
- 利用前门校正提供无偏的因果估计。
- 在真实世界对话数据集上展示有效性。
提出的方法
- 提出两智能体 MACR 框架:推理智能体基于保罗-埃尔德批判性思维与认知评估理论,生成四个子智能体的反事实中介变量 M。
- 实现一个带偏差感知的决策智能体,通过前门校正以 P(Y|do(X)) 的估计形式,利用 P(m|X) 与 P(Y|do(m))。
- 将前门校正分解为两部分计算,并由专门的智能体实现。
- 使用聚类(E5-large-v2 编码器 + K-means)将反事实推断整合为具有代表性的中介节点。
- 以采样为基础、分层的上下文内提示策略近似 P(Y|do(m)),以减轻分布偏倚。
- 与多种基线进行对比,包括传统风险模型、去偏语言模型、大型语言模型与图神经网络,在两个真实世界数据集上进行评估。
实验结果
研究问题
- RQ1在线自杀话语中不可观测的混淆因素是否可通过前门校正结合中介变量来处理?
- RQ2用反事实心理中介变量丰富上下文,是否能比仅使用日志交互来提升自杀风险预测?
- RQ3多智能体协作(推理+带偏见感知的决策)对预测性能与鲁棒性有何影响?
- RQ4在该领域,提出的前门方法是否优于标准后门去偏策略?
主要发现
- MACR 在两个数据集上达到最先进的加权 F1(PFA:0.3768;SCT:0.5108)。
- 在 SCT 数据集上,MACR 相较第二好模型在加权 F1 上提升 12.3%。
- 去偏语言模型有提升但未达到 MACR 的性能,凸显仅依赖后门去偏的局限性。
- 消融实验显示推理智能体对性能贡献最大,多智能体协作与前门校正均为关键。
- 基于前门的提示在此任务中优于上下文内学习与链式推理提示。
更好的研究,从现在开始
从阅读论文到最终审阅,大幅缩短您的研究时间。
无需绑定信用卡
本解读由 AI 生成,并经人工编辑审核。