paper_type.json (244B)
1 { 2 "paper_type": "empirical", 3 "reason": "The paper presents experimental results of a fine-tuned code LLM evaluated on standard benchmarks (HumanEval, CoderEval, LeetCode) with quantitative performance metrics as the primary contribution." 4 }