Schema: Discovering Unknown Environments via Agentic Program Induction

arXiv:2609.39140v1 Announce Type: new Abstract: Learning to complete tasks in unfamiliar environments with unknown rules remains a key challenge for LLM agents. Current LLM agents often record their discoveries in prose, which may not provide a compact, explicit account of how the environment works. Inspired by how scientists organize observations into testable, predictive theories, we introduce Schema, an agent harness that organizes learning and action through interactive program induction. The LLM agent decides what to investigate and how to act, expressing its evolving understanding of the environment as executable programs. The harness consists of a persistent program workspace and a small set of interfaces for checking these programs against the interaction history, planning within them, and executing plans under step-by-step verification. Schema raises ARC-AGI-3 RHAE from 58.7% to 99.2% with the same base model, solves 100% of the public DiG-bench games, and reaches the median performance of the top-50 human players on MazeBench. Extensive analysis shows the effectiveness of Schema in unknown mechanism discovery, and ablations confirm the contribution of each component.

Related

• Lifting Traces to Logic: Programmatic Skill Induction with Neuro-Symbolic Learning for Long-Horizon Agentic Tasks
• EnvCraft: Synthesizing Executable Environments in Agentic RL for Claw-like Agent
• Verifiable Hidden Dynamics Play: Generating Agentic RL Environments from Solved Mechanisms
• Trace2Tower: Transition-Aware EigenTrace Induction of Multi-Level Skills for LLM Agents

Source: arXiv cs.AI | 2026-10-01