ShortCoder: Knowledge-Augmented Syntax Optimization for Token-Efficient Code Generation

TL;DR

ShortCoder is a framework that enhances token efficiency in code generation by applying syntax simplification rules, a hybrid data synthesis pipeline, and fine-tuning strategies, leading to significant improvements without sacrificing code quality.

Contribution

It introduces syntax-level simplification rules, a hybrid data synthesis pipeline, and a fine-tuning strategy to improve token efficiency in code generation models.

Findings

Achieves 18.1%-37.8% token reduction in code generation.

Outperforms state-of-the-art methods on HumanEval.

Maintains code correctness and readability despite simplifications.

Abstract

Code generation tasks aim to automate the conversion of user requirements into executable code, significantly reducing manual development efforts and enhancing software productivity. The emergence of large language models (LLMs) has significantly advanced code generation, though their efficiency is still impacted by certain inherent architectural constraints. Each token generation necessitates a complete inference pass, requiring persistent retention of contextual information in memory and escalating resource consumption. While existing research prioritizes inference-phase optimizations such as prompt compression and model quantization, the generation phase remains underexplored. To tackle these challenges, we propose a knowledge-infused framework named ShortCoder, which optimizes code generation efficiency while preserving semantic equivalence and readability. In particular, we introduce: (1) ten syntax-level simplification rules for Python, derived from AST-preserving transformations, achieving 18.1% token reduction without functional compromise; (2) a hybrid data synthesis pipeline integrating rule-based rewriting with LLM-guided refinement, producing ShorterCodeBench, a corpus of validated tuples of original code and simplified code with semantic consistency; (3) a fine-tuning strategy that injects conciseness awareness into the base LLMs. Extensive experimental results demonstrate that ShortCoder consistently outperforms state-of-the-art methods on HumanEval, achieving an improvement of 18.1%-37.8% in generation efficiency over previous methods while ensuring the performance of code generation.