NaturalEdit: Code Modification through Direct Interaction with Adaptive Natural Language Representation

2025-10-06

AI Summary

Code modification requires integrated capabilities in comprehension, planning, expression, and verification—capabilities inadequately supported by static natural-language (NL) code summarization. This paper introduces IntenCode, an interactive NL summarization framework specifically designed for code modification tasks. Its core contributions are: (1) adaptive multi-granularity summary generation grounded in cognitive dimensions theory, enabling controllable abstraction-level adjustment; (2) fine-grained bidirectional mapping between source code and NL summaries to preserve structural and semantic fidelity; and (3) intent-driven real-time synchronized editing, substantially reducing edit latency and cognitive friction. Technical evaluation demonstrates IntenCode’s superiority over baselines in summary quality (BLEU+ROUGE) and synchronization accuracy. A user study with 24 developers shows statistically significant improvements in code understanding (+38%), sense of control (+42%), and modification confidence (+46%).

Technology Category

• Natural Language Processing: Summarization
• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Large Multimodal Models (LMMs)

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Practical large-scale studies of user experience
• Web Mining and Content Analysis: Normalization, clustering, classification, and summarization of Web text

Abstract

Code modification requires developers to comprehend code, plan changes, articulate intentions, and validate outcomes, making it a cognitively demanding process. Generated natural language code summaries aid comprehension but remain static and limited in supporting the full workflow. We present NaturalEdit, a system that makes code summaries interactive and adaptive representations directly linked to source code. Grounded in the Cognitive Dimensions of Notations, NaturalEdit implements a paradigm of code modification through interaction with natural language representations through three key features: (1) adaptive multi-faceted representation of code summaries with flexible Abstraction Gradient; (2) interactive mapping mechanisms between summaries and codes, ensuring a tight Closeness of Mapping; and (3) intent-driven, bidirectional synchronization that reduces Viscosity in editing and validation. A technical evaluation confirms the performance of NaturalEdit, and a user study with 12 developers shows that it enhances comprehension, intent articulation, and validation, giving developers greater confidence and control.

Problem

• Making code summaries interactive for direct modification
• Reducing cognitive demands in code comprehension and editing
• Enabling intent-driven bidirectional synchronization between code and summaries

Innovation

• Interactive natural language code summaries directly linked to source
• Adaptive multi-faceted representation with flexible abstraction gradient
• Bidirectional synchronization reducing viscosity in editing and validation