CROSSAGENTIE: Cross-Type and Cross-Task Multi-Agent LLM Collaboration for Zero-Shot Information Extraction

Large language models (LLMs) excel in generating unstructured text.However, they struggle with producing structured output while maintaining accuracy in zero-shot information extraction (IE), such as named entity recognition (NER) and relation extraction (RE).To address these challenges, we propose CROSSAGEN-TIE, a multi-agent framework that enhances zero-shot IE through multi-agent LLM collaboration.CROSSAGENTIE refines LLM predictions iteratively through two mechanisms: intra-group cross-type debate, which resolves entity-label conflicts through context-based evidence and confidence aggregation, and intergroup cross-task debate, where NER and RE mutually refine outputs via bidirectional feedback.Furthermore, we introduce template finetuning, distilling high-confidence multi-agent outputs into a single model, significantly reducing inference costs while preserving accuracy.Experiments across five NER and five RE datasets show that CROSSAGENTIE significantly outperforms state-of-the-art zero-shot baselines by a large margin.CROSSAGEN-TIE effectively addresses LLM limitations in structured prediction with an effective and efficient approach for zero-shot information extraction.Our GitHub can be found at https://github.com/Luca/CorssAgentIE.