Search papers, labs, and topics across Lattice.
AgentMercury introduces a scalable framework for synthesizing executable environments from high-level business scenarios, addressing the limitations of manually constructed task-centric environments. By creating 4,783 diverse environments across various industries and countries, the framework enables the emergence of complex tasks and interaction trajectories. Policies trained on these environments significantly outperform traditional benchmarks, with notable improvements in enterprise workflows and other domains, demonstrating the potential for generalizable learning signals in reinforcement learning.
Training on AgentMercury's synthesized environments boosts policy performance by over 25% on enterprise workflows, revealing a new frontier in scalable RL training.
Agents learn to act through interaction with environments, yet the environments used for training are often manually constructed or synthesized around predefined tasks and benchmarks. This task-centric paradigm makes it difficult to scale environments that reflect realistic and evolving workflows where diverse tasks can naturally emerge from the underlying world. We introduce AgentMercury, a scalable framework for synthesizing executable environments from high-level business scenarios. Rather than constructing an environment for a specific task, AgentMercury first instantiates a persistent world with entities, services, tools, state, and executable cross-service invariants, from which diverse tasks and interaction trajectories can subsequently emerge. We construct 4,783 executable environments spanning 14 industries and 50 countries, and use them as training substrates for reinforcement learning. Despite being generated without targeting the evaluation benchmarks, policies trained on these business-oriented environments improve substantially on both enterprise workflows and out-of-domain benchmarks spanning reasoning, coding, scientific computing, and tool use. In our experiments, Qwen3.5-4B improves from 12.3 to 15.7 on EnterpriseOps-GYM and from 45.9 to 56.0 on AIME26 after training on AgentMercury environments. We further show that the construction process itself can be learned: fine-tuning Qwen3.5-35B-A3B on construction traces increases executable-world authoring success from 3.3% to 83.3% on held-out business scenarios. These results show that scenario-grounded environments can provide useful and generalizable learning signals beyond benchmark-specific training, while their construction can itself become a learnable capability.
