Rethinking IDE Customization for Enhanced HAX: A Hyperdimensional Perspective

TL;DR

The paper addresses mismatches between developer preferences and AI-generated code in IDEs by proposing Hyper-Dimensional Computing (HDC) as a compact, context-aware representation in $D$-dimensional spaces. It introduces the MAP framework with binding $A ⊗ B$, bundling $A ⊕ B$, and permutation $P$ to encode user actions, stylistic preferences, and project context, including concrete encodings for sequences, style mappings, and project settings. The authors outline three use cases—action-sequence encoding for next-action prediction, style-matched generation, and project-context encoding—to illustrate practical HAX personalization. The work highlights the potential of HD representations to enable efficient, low-overhead customization in AI-enhanced development environments and motivates further integration into IDE toolchains.

Abstract

As Integrated Development Environments (IDEs) increasingly integrate Artificial Intelligence, Software Engineering faces both benefits like productivity gains and challenges like mismatched user preferences. We propose Hyper-Dimensional (HD) vector spaces to model Human-Computer Interaction, focusing on user actions, stylistic preferences, and project context. These contributions aim to inspire further research on applying HD computing in IDE design.