Agents

When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents

Single source1 source12 Oct 2026

arXiv:2605.11928v2 Announce Type: replace Abstract: Tool-use language agents are evaluated on benchmarks that assume clean inputs, unambiguous tool registries, and reliable APIs.

Reported by 1 source

arxiv.orghttps://arxiv.org/abs/2605.11928

← Back to all developments · Signals are a development reported by more than one source. See the methodology.