这项研究揭示了LLM在立场检测中的系统性错误模式，对做NLP评估和模型优化的团队有直接参考价值——SICI指数可以帮你快速识别模型在哪些样本上会失效，建议做立场检测或模型鲁棒性研究的点开看看。
研究者提出SICI（立场推理复杂度指数），一个七维诊断指标，用于衡量目标-文本对在语义和语用上的复杂度。该指数在SemEval-2016和VAST数据集上比表面代理指标更好地预测LLM的准确性，且具有较高的跨评分者信度（α=0.771）。关键发现是，随着SICI增加，LLM错误模式发生阶段转变：低复杂度样本易导致过度归因（尤其是反对立场），中等复杂度样本形成不稳定边界，高复杂度样本则快速集中到“无立场”预测。这种结构在GPT-3.5、GPT-4o-mini、DeepSeek-V3和GPT-4o中一致存在，但更强模型会移动边界。15种干预方法的实验表明，提示、检索和辩论往往只是沿归因-弃权轴移动模型，而非消除高复杂度的瓶颈。
SICI: A Semantic-Pragmatic Complexity Index Reveals Regime Shifts in LLM Stance Detection
Prompt-based LLMs are increasingly used for stance detection, but harder examples are not always repaired by clearer instructions, reasoning prompts, retrieval, or debate. We introduce SICI (Stance Inference Complexity Index), a seven-dimensional diagnostic measure of the semantic-pragmatic burden imposed by a target--text pair. Across SemEval-2016 and VAST, SICI predicts LLM accuracy better than surface proxies and shows substantial cross-scorer reliability ($α=0.771$). More importantly, LLM errors change regime as SICI increases: low-complexity examples invite over-attribution, especially Against predictions; intermediate examples form an unstable boundary; and high-complexity examples rapidly concentrate on None. This phase-transition-like structure persists across GPT-3.5, GPT-4o-mini, DeepSeek-V3, and GPT-4o, although stronger models move the boundaries. A 15-method intervention study further shows that prompting, retrieval, and debate often shift models along the attribution--abstention axis rather than removing the high-complexity bottleneck.