通过多次运行模型并取一致结果，提升情感分析准确性。
nchellwig at SemEval-2026 Task 3: Self-Consistent Structured Generation (SCSG) for Dimensional Aspect-Based Sentiment Analysis using Large Language Models
- 对同一输入多次运行大模型，只保留多数一致的结果
- 15次运行后在6语言8场景中显著优于单次推理
- 适合需要高可靠性的跨语言情感分析任务
我们为 SemEval-2026 Task 3（Track A）提出自一致性结构生成（SCSG）方法，用于维度化方面情感分析。SCSG 通过多次执行经 LoRA 微调的大语言模型（每实例 15 次），仅保留多轮推理中达成多数共识的三元组，以增强预测可靠性。为缓解多次前向传播带来的计算开销，采用 vLLM 的 PagedAttention 机制实现键值缓存高效复用。在 6 种语言和 8 个语言-领域组合上的评估表明，该方法相比单次推理提示显著提升性能；基于 Gemma 3 的系统在所有设置中均位列前七，其中在三个英文子集上获第二名，在鞑靼语-餐饮领域（Tatar-Restaurant）取得第一名。
原文摘要 · Abstract (English)
We present Self-Consistent Structured Generation (SCSG) for Dimensional Aspect-Based Sentiment Analysis in SemEval-2026 Task 3 (Track A). SCSG enhances prediction reliability by executing a LoRA-adapted large language model multiple times per instance, retaining only tuples that achieve a majority consensus across runs. To mitigate the computational overhead of multiple forward passes, we leverage vLLM's PagedAttention mechanism for efficient key--value cache reuse. Evaluation across 6 languages and 8 language--domain combinations demonstrates that self-consistency with 15 executions yields statistically significant improvements over single-inference prompting, with our system (leveraging Gemma 3) ranking in the top seven across all settings, achieving second place on three out of four English subsets and first place on Tatar-Restaurant for DimASTE.
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。