Code2LoRA: Hypernetwork Generates Repo-Specific Adapters for Code LMs with Zero Inference Overhead
University of Waterloo
Code2LoRA generates repository-specific LoRA adapters for code language models with zero inference-time token overhead. Two variants: Code2LoRA-Static converts a repo snapshot into an adapter; Code2LoRA-Evo maintains adapters via GRU state updated per code diff. Introduces RepoPeftBench (604 Python repos, static and evolution tracks). Code2LoRA-Static achieves 63.8% cross-repo and 66.2% in-repo exact match, matching per-repository LoRA fine-tuning without any per-repo training.
Why it matters
Addresses a practical bottleneck for code AI in production: keeping LLM adapters up to date as codebases evolve without re-running expensive fine-tuning. The GRU-based incremental update mechanism enables adapter maintenance at software-evolution speed.
Importance: 2/5
Official arXiv + HuggingFace; practical code LM adaptation technique with production relevance and new benchmark.