Benchmarking Agentic Systems in Automated Scientific Information Extraction with ChemX
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866911187767132160 |
|---|---|
| author | Vepreva, Anastasia Razlivina, Julia Eremeeva, Maria Gubina, Nina Orlova, Anastasia Dmitrenko, Aleksei Kapranova, Ksenya Jyakhwo, Susan Vasilev, Nikita Sarkisyan, Arsen Chernyshov, Ivan Yu. Vinogradov, Vladimir Dmitrenko, Andrei |
| author_facet | Vepreva, Anastasia Razlivina, Julia Eremeeva, Maria Gubina, Nina Orlova, Anastasia Dmitrenko, Aleksei Kapranova, Ksenya Jyakhwo, Susan Vasilev, Nikita Sarkisyan, Arsen Chernyshov, Ivan Yu. Vinogradov, Vladimir Dmitrenko, Andrei |
| contents | The emergence of agent-based systems represents a significant advancement in artificial intelligence, with growing applications in automated data extraction. However, chemical information extraction remains a formidable challenge due to the inherent heterogeneity of chemical data. Current agent-based approaches, both general-purpose and domain-specific, exhibit limited performance in this domain. To address this gap, we present ChemX, a comprehensive collection of 10 manually curated and domain-expert-validated datasets focusing on nanomaterials and small molecules. These datasets are designed to rigorously evaluate and enhance automated extraction methodologies in chemistry. To demonstrate their utility, we conduct an extensive benchmarking study comparing existing state-of-the-art agentic systems such as ChatGPT Agent and chemical-specific data extraction agents. Additionally, we introduce our own single-agent approach that enables precise control over document preprocessing prior to extraction. We further evaluate the performance of modern baselines, such as GPT-5 and GPT-5 Thinking, to compare their capabilities with agentic approaches. Our empirical findings reveal persistent challenges in chemical information extraction, particularly in processing domain-specific terminology, complex tabular and schematic representations, and context-dependent ambiguities. The ChemX benchmark serves as a critical resource for advancing automated information extraction in chemistry, challenging the generalization capabilities of existing methods, and providing valuable insights into effective evaluation strategies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_00795 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Benchmarking Agentic Systems in Automated Scientific Information Extraction with ChemX Vepreva, Anastasia Razlivina, Julia Eremeeva, Maria Gubina, Nina Orlova, Anastasia Dmitrenko, Aleksei Kapranova, Ksenya Jyakhwo, Susan Vasilev, Nikita Sarkisyan, Arsen Chernyshov, Ivan Yu. Vinogradov, Vladimir Dmitrenko, Andrei Artificial Intelligence The emergence of agent-based systems represents a significant advancement in artificial intelligence, with growing applications in automated data extraction. However, chemical information extraction remains a formidable challenge due to the inherent heterogeneity of chemical data. Current agent-based approaches, both general-purpose and domain-specific, exhibit limited performance in this domain. To address this gap, we present ChemX, a comprehensive collection of 10 manually curated and domain-expert-validated datasets focusing on nanomaterials and small molecules. These datasets are designed to rigorously evaluate and enhance automated extraction methodologies in chemistry. To demonstrate their utility, we conduct an extensive benchmarking study comparing existing state-of-the-art agentic systems such as ChatGPT Agent and chemical-specific data extraction agents. Additionally, we introduce our own single-agent approach that enables precise control over document preprocessing prior to extraction. We further evaluate the performance of modern baselines, such as GPT-5 and GPT-5 Thinking, to compare their capabilities with agentic approaches. Our empirical findings reveal persistent challenges in chemical information extraction, particularly in processing domain-specific terminology, complex tabular and schematic representations, and context-dependent ambiguities. The ChemX benchmark serves as a critical resource for advancing automated information extraction in chemistry, challenging the generalization capabilities of existing methods, and providing valuable insights into effective evaluation strategies. |
| title | Benchmarking Agentic Systems in Automated Scientific Information Extraction with ChemX |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2510.00795 |