paper_type.json (257B)
1 { 2 "paper_type": "empirical", 3 "reason": "Reports quantitative experimental results on benchmarks (VerilogEval, RTLLM) with a fine-tuned model, including ablation studies validating that functional correctness in training data improves LLM performance." 4 }