We introduce the Universal Document-level Information Extraction (DocIE) challenge on XLLM platform. The goal of LLM-SR is to generate a controllable and interpretable reasoning process by leveraging structural reasoning. This challenge requires the structural parsing of two distinct components: major premises and minor premises, then involving identifying fine-grained “alignments” between these two structures and ultimately deriving a conclusion. This task can be regarded as a constrained Chain-of-Thought (CoT) reasoning process, where reasoning is conducted step by step with reference to facts and relevant rules, thereby improving the transparency and reliability of the process.
LLMSR Challenge and Dataset is coming soon.
Please refer to the challenge webpage: LLMSR Challenge