Semantic role labeling (SRL) enriches many downstream applications, e.g., machine translation, question answering, summarization, and stance/belief detection. However, building multilingual SRL models is challenging due to the scarcity of semantically annotated corpora for multiple languages. Moreover, state-of-the-art SRL projection (XSRL) based on large language models (LLMs) yields output that is riddled with spurious role labels. Remediation of such hallucinations is not straightforward due to the lack of explainability of LLMs. We show that hallucinated role labels are related to naturally occurring divergence types that interfere with initial alignments. We implement Divergence-Aware Hallucination-Remediated SRL projection (DAHRS), leveraging linguistically-informed alignment remediation followed by greedy First-Come First-Assign (FCFA) SRL projection. DAHRS improves the accuracy of SRL projection without additional transformer-based machinery, beating XSRL in both human and automatic comparisons, and advancing beyond headwords to accommodate phrase-level SRL projection (e.g., EN-FR, EN-ES). Using CoNLL-2009 as our ground truth, we achieve a higher word-level F1 over XSRL: 87.6% vs. 77.3% (EN-FR) and 89.0% vs. 82.7% (EN-ES). Human phrase-level assessments yield 89.1% (EN-FR) and 91.0% (EN-ES). We also define a divergence metric to adapt our approach to other language pairs (e.g., English-Tagalog).

翻译：语义角色标注（SRL）为许多下游应用（如机器翻译、问答、摘要以及立场/信念检测）提供了丰富信息。然而，由于多语言语义标注语料稀缺，构建多语言SRL模型颇具挑战。此外，基于大语言模型（LLM）的最先进SRL投影方法（XSRL）会产生大量虚假角色标签。由于LLM缺乏可解释性，修正此类幻觉并非易事。我们证明，幻觉角色标签与干扰初始对齐的自然发散类型相关。我们实现了基于发散感知的幻觉修正SRL投影（DAHRS），该方法利用语言学启发的对齐修正，随后进行贪心的先到先分配（FCFA）SRL投影。DAHRS在不增加基于Transformer架构的情况下提升了SRL投影的准确性，在人工与自动评估中均优于XSRL，并超越词元层面以支持短语级SRL投影（如英-法、英-西）。以CoNLL-2009作为基准，我们在词级F1上取得了优于XSRL的结果：87.6% 对比 77.3%（英-法）以及89.0% 对比 82.7%（英-西）。人工短语级评估结果为89.1%（英-法）和91.0%（英-西）。我们还定义了一种发散度量指标，使我们的方法能够适配其他语言对（如英语-他加禄语）。