English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Large Language Models (LLMs) have demonstrated impressive inferentialcapabilities, with numerous research endeavors devoted to enhancing thiscapacity through prompting. Despite these efforts, a unified epistemologicalfoundation is still conspicuously absent. Drawing inspiration from Kant's apriori philosophy, we propose the UPAR prompting framework, designed to emulatethe structure of human cognition within LLMs. The UPAR framework is delineatedinto four phases: "Understand", "Plan", "Act", and "Reflect", enabling theextraction of structured information from complex contexts, prior planning ofsolutions, execution according to plan, and self-reflection. This structuresignificantly augments the explainability and accuracy of LLM inference,producing a human-understandable and inspectable inferential trajectory.Furthermore, our work offers an epistemological foundation for existingprompting techniques, allowing for a possible systematic integration of thesemethods. With GPT-4, our approach elevates the accuracy from COT baseline of22.92% to 58.33% in a challenging subset of GSM8K, and from 67.91% to 75.40% inthe causal judgment task. Without using few-shot examples or external tools,UPAR significantly outperforms existing prompting methods on SCIBENCH, achallenging dataset containing collegiate-level mathematics, chemistry, andphysics scientific problems.

UPAR: A Kantian-Inspired Prompting Framework for Enhancing Large Language Model Capabilities