MoleculeQA: A Dataset to Evaluate Factual Accuracy in Molecular Comprehension
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914711754244096 |
|---|---|
| author | Lu, Xingyu Cao, He Liu, Zijing Bai, Shengyuan Chen, Leqing Yao, Yuan Zheng, Hai-Tao Li, Yu |
| author_facet | Lu, Xingyu Cao, He Liu, Zijing Bai, Shengyuan Chen, Leqing Yao, Yuan Zheng, Hai-Tao Li, Yu |
| contents | Large language models are playing an increasingly significant role in molecular research, yet existing models often generate erroneous information, posing challenges to accurate molecular comprehension. Traditional evaluation metrics for generated content fail to assess a model's accuracy in molecular understanding. To rectify the absence of factual evaluation, we present MoleculeQA, a novel question answering (QA) dataset which possesses 62K QA pairs over 23K molecules. Each QA pair, composed of a manual question, a positive option and three negative options, has consistent semantics with a molecular description from authoritative molecular corpus. MoleculeQA is not only the first benchmark for molecular factual bias evaluation but also the largest QA dataset for molecular research. A comprehensive evaluation on MoleculeQA for existing molecular LLMs exposes their deficiencies in specific areas and pinpoints several particularly crucial factors for molecular understanding. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_08192 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | MoleculeQA: A Dataset to Evaluate Factual Accuracy in Molecular Comprehension Lu, Xingyu Cao, He Liu, Zijing Bai, Shengyuan Chen, Leqing Yao, Yuan Zheng, Hai-Tao Li, Yu Computation and Language Biomolecules Large language models are playing an increasingly significant role in molecular research, yet existing models often generate erroneous information, posing challenges to accurate molecular comprehension. Traditional evaluation metrics for generated content fail to assess a model's accuracy in molecular understanding. To rectify the absence of factual evaluation, we present MoleculeQA, a novel question answering (QA) dataset which possesses 62K QA pairs over 23K molecules. Each QA pair, composed of a manual question, a positive option and three negative options, has consistent semantics with a molecular description from authoritative molecular corpus. MoleculeQA is not only the first benchmark for molecular factual bias evaluation but also the largest QA dataset for molecular research. A comprehensive evaluation on MoleculeQA for existing molecular LLMs exposes their deficiencies in specific areas and pinpoints several particularly crucial factors for molecular understanding. |
| title | MoleculeQA: A Dataset to Evaluate Factual Accuracy in Molecular Comprehension |
| topic | Computation and Language Biomolecules |
| url | https://arxiv.org/abs/2403.08192 |