TicTacBench：评估编程智能体 RTL 时序收敛能力的基准
TicTacBench: Benchmarking Timing Closure Capabilities of Coding Agents
AI 写代码很溜，但芯片时序收敛还是硬骨头：8 个前沿 LLM 跑了 300 多次，最强的也只过了 53.3% 的任务。
研究团队发布 TicTacBench，用 30 个任务测试编程智能体在 RTL 时序收敛上的能力，每个任务附带次优 RTL 设计、真实时序约束、功能等价性验证和时序报告，并以布局布线后（post-PnR）结果判定。
在超过 300 次运行、覆盖 8 个前沿 LLM 的测试中，最好的智能体也只完成 53.3% 的任务，平均造成 7.
18% 的 ADP（面积延迟积）劣化，同时带来 8.83% 的 EDDP 改善。
论文归纳出智能体时序收敛失败的常见类别，并提出 TicTacSkill 方法，引导智能体遵循标准时序收敛流程，将时序收敛率提高 9%。
TicTacBench: Benchmarking Timing Closure Capabilities of Coding Agents
Recent advances in large language models (LLMs) have led to the emergence of coding agents capable of performing complex engineering tasks, including register-transfer level (RTL) design and optimization. Existing RTL benchmarks mainly evaluate functional correctness and performance, power, and area (PPA) of the generated RTL designs, leaving agents' ability for \emph{timing closure} under-evaluated. We propose TicTacBench, a benchmark specifically designed to evaluate coding agents' capabilities for RTL-level timing closure under post-place-and-route (post-PnR) evaluation. TicTacBench contains 30 diverse tasks, each provided with a suboptimal RTL design, realistic timing constraints, functional equivalence verification, and timing reports. With over 300 runs of coding agents driven by 8 frontier LLMs, we find that even the best agent can only close 53.3\% of tasks with 7.18\% area-delay product (ADP) degradation and 8.83\% energy-delay-squared product (EDDP) improvement on average. We identify common failure categories that explain why agents fail to close timing. Then we propose TicTacSkill, a new method that guides agents to follow standard timing-closure procedures and improves the Timing Closure Rate by 9\%. These results suggest that while coding agents have made significant progress in RTL design, their timing-closure capability still has substantial room for improvement.
This report is categorized under AITOP's "Papers" desk. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include TicTacBench, TicTacSkill, coding assistants, agents.
Source attribution: curated from arXiv cs.AI (@Bowei Wang, Zhigang Fang, Zhijie Yang, Renzhi Chen, Shanshan Li, Lei Wang).
