TRACE: Evaluating Execution Efficiency of LLM-Based Code Translation

While Large Language Models (LLMs) have substantially improved the functional correctness of code translation, the critical dimension of $\mathit{\text{execution}}\mathit{\text{efficiency}}$ remains overlooked. We present $\mathbf{\text{textsc}}\mathbf{\text{trace}}$, the first benchmark to explicitly assess efficiency in LLM-translated code. $\text{textsc}trace$ includes 1,000 efficiency-critical tasks across C++, Java, and Python, each augmented with stress tests that reveal efficiency degradations often overlooked by small-scale tests. Using $\text{textsc}trace$, we conduct an extensive evaluation of 28 representative LLMs and highlight several key insights: 1) Correctness is not a reliable proxy for efficiency: the correctness leader $\mathit{\text{Claude-4-think}}$ achieves only mid-level time efficiency, outperformed by smaller open-source LLMs such as $\mathit{\text{Qwen2.5-Coder-14B-Instruct}}$. 2) Inefficiency is both prevalent and patterned: 23.5% of correct translations exhibit pronounced inefficiency, distributed across algorithmic faults (11.9%), language construct mismatches (66.4%), and resource mismanagement (21.7%). 3) Inference-time prompt strategies bring only modest improvements, suggesting that current LLMs lack intrinsic efficiency awareness. Together, our results establish efficiency as an essential dimension of code translation and position $\text{textsc}trace$ as a principled foundation for efficiency-oriented evaluation.