Search papers, labs, and topics across Lattice.
This paper introduces AgentMercury, a scalable framework that synthesizes executable environments from high-level business scenarios, moving beyond the limitations of task-centric environment construction. By creating 4,783 diverse environments across various industries and countries, the framework enables reinforcement learning agents to learn from realistic workflows and emergent tasks. The results demonstrate significant performance improvements in enterprise workflows and out-of-domain benchmarks, highlighting the framework's effectiveness in generating generalizable learning signals and its potential for automating environment construction through learned processes.
Training on environments synthesized from real-world business scenarios boosts agent performance on both enterprise tasks and diverse benchmarks, 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.
