Metraipolis › archive › dc_e87c63e4b062

MatToolBench: Benchmarking Multimodal Agents in Real-World Materials Science Workflows — arXiv:2609.37053

Source
arXiv (en)
Original
https://arxiv.org/abs/2609.37053
License (as archived)
CC0-1.0
Archived
2026-10-07T16:23:27.218Z
SHA-256 of the archived text
e1ddfb943ebc5a22ba2a1b15085c05d0e313724512f799ad27baa4a435c004c4
Stable URL of this page
https://metraipolis.com/documento/dc_e87c63e4b062

How to cite

MatToolBench: Benchmarking Multimodal Agents in Real-World Materials Science Workflows — arXiv:2609.37053 — arXiv (en), https://arxiv.org/abs/2609.37053 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e87c63e4b062 (Metraipolis archive dc_e87c63e4b062, sha256 e1ddfb943ebc5a22ba2a1b15085c05d0e313724512f799ad27baa4a435c004c4)
BibTeX
@misc{dce87c63e4b062,
  title = {MatToolBench: Benchmarking Multimodal Agents in Real-World Materials Science Workflows — arXiv:2609.37053},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.37053},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e87c63e4b062; sha256 e1ddfb943ebc5a22ba2a1b15085c05d0e313724512f799ad27baa4a435c004c4},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.37053. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multimodal GUI agents have achieved impressive results on general software benchmarks, yet their ability to operate professional scientific software remains largely unexplored. In materials science, sparse domain-specific web data, specialized interfaces, and tacit workflow conventions create blind spots that general-purpose pretraining cannot readily bridge. We present MatToolBench, the first real-environment benchmark for evaluating multimodal GUI agents on professional materials science software, comprising 204 tasks across 10 tools in three modalities: GUI operation, OriginPro scripting, and code-based database queries, all executed inside a Windows 11 VM. Each task is decomposed into fine-grained sub-criteria by domain experts, enabling interpretable partial-credit scoring; the GUI component of our multi-level evaluation pipeline achieves an average F1 of 0.98. For OriginPro figure-generation tasks, we further conduct a human-LLM agreement study to validate the use of a multimodal judge for secondary aesthetic assessment. Our experiments show that strong performance on general benchmarks does not transfer to professional scientific workflows, and that this gap is not a visual-grounding problem alone: failures arise from domain-specific operational knowledge, sparse pretraining coverage of scientific software, weak cross-tool artifact handoff, and critical states exposed only visually. Even the best model reaches only 25% success rate on GUI tasks and 45% on code tasks. MatToolBench therefore serves as a challenging diagnostic benchmark and real-environment testbed for data-scarce, knowledge-intensive scientific workflows.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-30 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2609.37053 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito).
Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina.
Letto il 2026-10-07T15:54:53.584Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-09-30?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).