mlsanityData to 7 Oct 2026

ChemWorld: Programmable Chemical Worlds for Controlled and Replayable Agent Experimentation

arXiv 2608.10792 ↗First posted 11 Aug 2026cs.AI cs.LGcited by 0

Not on the map. Why?

Abstract (arXiv)

Autonomous chemistry increasingly depends on environments in which agents can repeatedly act, observe, and adapt.Physical laboratories provide essential real-material evidence but are costly to repeat and difficult to use for tightly matched interventions, whereas most digital environments keep the underlying experimental world largely fixed. We introduce ChemWorld, a programmable chemical environment in which reusable process and observation components are compiled into executable worlds. ChemWorld separates the public experimental contract available to an agent from evaluator-owned chemical and material laws. Researchers can therefore vary world composition and operating conditions, or change a single hidden law while holding the public task and interaction conditions fixed. Transactional execution records operations, failures, resource changes, and state transitions, allowing complete environment-action trajectories to be replayed exactly and audited. Full-census qualification covered the reference registry, 52 generated compositions, and module, interface, compilation, and invalid-action tests. Eight deterministic experimental cases demonstrated shared lifecycle semantics, failure recovery, and exact replay, while six parent-child world-fork pairs isolated the effects of single private-law interventions under matched public conditions. An independent agent also completed a full lifecycle in a non-reference world through the same public interface. Within the declared component and model domain, ChemWorld provides a controlled and replayable substrate for studying experimentation across systematically varied chemical worlds, complementary to physical-laboratory evidence and calibration.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Builds on 14 of its 22 references that are in the map. First posted on arXiv 11 Aug 2026.

Builds on

14 of 22 references.

References · 14

  1. 2023ChemCrow: Augmenting large-language models with chemistry toolscited by 609
  2. 2024DISCOVERYWORLD: A Virtual Environment for Developing and Evaluating Automated Scientific Discovery Agentscited by 89
  3. 2024ORGANA: A Robotic Assistant for Automated Chemistry Experimentation and Characterizationcited by 60
  4. 2025NewtonBench: Benchmarking Generalizable Scientific Law Discovery in LLM Agentscited by 32
  5. 2020Olympus: a benchmarking framework for noisy optimization and experiment planningcited by 28
  6. 2025BoxingGym: Benchmarking Progress in Automated Experimental Design and Model Discoverycited by 15
  7. 2025LabUtopia: High-Fidelity Simulation and Hierarchical Benchmark for Scientific Embodied Agentscited by 14
  8. 2025AutoLabs: Cognitive Multi-Agent Systems with Self-Correction for Autonomous Chemical Experimentationcited by 13
  9. 2025Measuring Scientific Capabilities of Language Models with a Systems Biology Dry Labcited by 11
  10. 2024PC-Gym: Benchmark Environments For Process Control Problemscited by 8
  11. 2025Agentic Exploration of Physics Modelscited by 5
  12. 2026MADE: Benchmark Environments for Closed-Loop Materials Discoverycited by 5
2 more
  1. 2026Scaling Scientific Discovery Environments for Turn-Level Agentic RLcited by 2
  2. 2026Labimus: A Simulation and Benchmark for Humanoid Dexterous Manipulation in Chemical Laboratorycited by 1

Built on by

Not cited yet.