CIRCUIT: A Benchmark for Circuit Interpretation and Reasoning Capabilities of LLMs
This provides a domain-specific benchmark for evaluating LLMs in analog circuit design, highlighting their limitations for researchers and engineers in electronics.
The authors tackled the lack of benchmarks for assessing LLMs' reasoning capabilities in analog circuit design by creating the CIRCUIT dataset with 510 question-answer pairs, where the best model, GPT-4o, achieved 48.04% accuracy on final answers and passed only 27.45% of unit tests.
The role of Large Language Models (LLMs) has not been extensively explored in analog circuit design, which could benefit from a reasoning-based approach that transcends traditional optimization techniques. In particular, despite their growing relevance, there are no benchmarks to assess LLMs' reasoning capability about circuits. Therefore, we created the CIRCUIT dataset consisting of 510 question-answer pairs spanning various levels of analog-circuit-related subjects. The best-performing model on our dataset, GPT-4o, achieves 48.04% accuracy when evaluated on the final numerical answer. To evaluate the robustness of LLMs on our dataset, we introduced a unique feature that enables unit-test-like evaluation by grouping questions into unit tests. In this case, GPT-4o can only pass 27.45% of the unit tests, highlighting that the most advanced LLMs still struggle with understanding circuits, which requires multi-level reasoning, particularly when involving circuit topologies. This circuit-specific benchmark highlights LLMs' limitations, offering valuable insights for advancing their application in analog integrated circuit design.