Open access
2026
MultiCAT-Bench: A Multi-Categorical Agent Tool Use Benchmark
MultiCAT-Bench, the first benchmark focused on detailed categorization of tasks for assessing tool utilization, is introduced, along with an approach for its automated generation, which utilizes the GPT-5 family.
A. Vyatkin, A. Poptsov, V. Oliseenko et al.
· IEEE Access · 0 citations