IE as Cache: Information Extraction Enhanced Agentic Reasoning
For researchers in information extraction and LLM reasoning, this work shows that IE can be reused as a dynamic cognitive resource to improve multi-step inference, offering a new perspective on downstream uses of IE.
The paper proposes IE-as-Cache, a framework that repurposes information extraction as a cognitive cache to enhance agentic reasoning, achieving significant improvements in reasoning accuracy across diverse LLMs on challenging benchmarks.
Information Extraction aims to distill structured, decision-relevant information from unstructured text, serving as a foundation for downstream understanding and reasoning. However, it is traditionally treated merely as a terminal objective: once extracted, the resulting structure is often consumed in isolation rather than maintained and reused during multi-step inference. Moving beyond this, we propose \textit{IE-as-Cache}, a framework that repurposes IE as a cognitive cache to enhance agentic reasoning. Drawing inspiration from hierarchical computer memory, our approach combines query-driven extraction with cache-aware reasoning to dynamically maintain compact intermediate information and filter noise. Experiments on challenging benchmarks across diverse LLMs demonstrate significant improvements in reasoning accuracy, indicating that IE can be effectively repurposed as a reusable cognitive resource and offering a promising direction for future research on downstream uses of IE.