KOCO-BENCH: Can Large Language Models Leverage Domain Knowledge in Software Development?

TL;DR

KOCO-BENCH is a new benchmark for evaluating how well large language models can learn and apply domain-specific knowledge in software development across multiple specialized domains.

Contribution

It introduces KOCO-BENCH, a comprehensive benchmark with knowledge corpora and multi-level tasks, to assess domain specialization methods for LLMs in real-world software projects.

Findings

State-of-the-art LLMs struggle with domain-specific tasks.

Existing domain specialization methods yield only marginal improvements.

The best model achieves just 34.2% accuracy, indicating significant challenges.

Abstract

Large language models (LLMs) excel at general programming but struggle with domain-specific software development, necessitating domain specialization methods for LLMs to learn and utilize domain knowledge and data. However, existing domain-specific code benchmarks cannot evaluate the effectiveness of domain specialization methods, which focus on assessing what knowledge LLMs possess rather than how they acquire and apply new knowledge, lacking explicit knowledge corpora for developing domain specialization methods. To this end, we present KOCO-BENCH, a novel benchmark designed for evaluating domain specialization methods in real-world software development. KOCO-BENCH contains 6 emerging domains with 11 software frameworks and 25 projects, featuring curated knowledge corpora alongside multi-granularity evaluation tasks including domain code generation (from function-level to project-level with rigorous test suites) and domain knowledge understanding (via multiple-choice Q&A). Unlike previous benchmarks that only provide test sets for direct evaluation, KOCO-BENCH requires acquiring and applying diverse domain knowledge (APIs, rules, constraints, etc.) from knowledge corpora to solve evaluation tasks. Our evaluations reveal that KOCO-BENCH poses significant challenges to state-of-the-art LLMs. Even with domain specialization methods (e.g., SFT, RAG, kNN-LM) applied, improvements remain marginal. Best-performing coding agent, Claude Code, achieves only 34.2%, highlighting the urgent need for more effective domain specialization methods. We release KOCO-BENCH, evaluation code, and baselines to advance further research at https://github.com/jiangxxxue/KOCO-bench.