Problem
Research questions and friction points this paper is trying to address.
causal discovery
text generation
causal graph annotation
naturalness
ground truth
iTAG: Inverse Design for Natural Text Generation with Accurate Causal Graph Annotations
🤖 AI Summary
This work addresses a critical limitation in textual causal discovery research—the scarcity of datasets that simultaneously exhibit natural language fluency and accurately annotated causal graphs. To overcome this, the authors propose a novel inverse-design framework that aligns high-fidelity causal structures with semantically coherent text while preserving linguistic naturalness. The approach assigns real-world concepts to nodes in a given causal graph and leverages large language models combined with chain-of-thought (CoT) reasoning to iteratively refine concept selection, thereby transforming structured causal graphs into fluent, contextually meaningful narratives. The resulting synthetic data achieves strong performance in both textual fluency and causal annotation accuracy, enabling causal discovery algorithms trained on it to match the performance observed on real-world data, thus offering a high-quality benchmark resource for the field.
📝 Abstract
A fundamental obstacle to causal discovery from text is the lack of causally annotated text data for use as ground truth, due to high annotation costs. This motivates an important task of generating text with causal graph annotations. Early template-based generation methods sacrifice text naturalness in exchange for high causal graph annotation accuracy. Recent Large Language Model (LLM)-dependent methods directly generate natural text from target graphs through LLMs, but do not guarantee causal graph annotation accuracy. Therefore, we propose iTAG, which performs real-world concept assignment to nodes before converting causal graphs into text in existing LLM-dependent methods. iTAG frames this process as an inverse problem with the causal graph as the target, iteratively examining and refining concept selection through Chain-of-Thought (CoT) reasoning so that the induced relations between concepts are as consistent as possible with the target causal relationships described by the causal graph. iTAG demonstrates both extremely high annotation accuracy and naturalness across extensive tests, and the results of testing text-based causal discovery algorithms with the generated data show high statistical correlation with real-world data. This suggests that iTAG-generated data can serve as a practical surrogate for scalable benchmarking of text-based causal discovery algorithms.
Innovation
Methods, ideas, or system contributions that make the work stand out.
inverse design
causal graph annotation
natural text generation
Chain-of-Thought reasoning
text-based causal discovery
🔎 Similar Papers
No similar papers found.
