Can LLMs Prove Robotic Path Planning Optimality? A Benchmark for Research-Level Algorithm Verification
Published in IEEE IROS 2026, 2026
This work introduces a benchmark of 34 research-level proof tasks for robotic path planning and studies how domain-specific context helps LLMs verify approximation guarantees.
Recommended citation: Yang, Z., Tazwar, M. T., Wei, M., & Zhu, Z. (2026). Can LLMs Prove Robotic Path Planning Optimality? A Benchmark for Research-Level Algorithm Verification. IEEE IROS 2026.