Mol-R1: Towards Explicit Long-CoT Reasoning in Molecule Discovery
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866908485022646272 |
|---|---|
| author | Li, Jiatong Wang, Weida Zhang, Qinggang Li, Junxian Zhang, Di Zheng, Changmeng Zhang, Shufei Wei, Xiaoyong Li, Qing |
| author_facet | Li, Jiatong Wang, Weida Zhang, Qinggang Li, Junxian Zhang, Di Zheng, Changmeng Zhang, Shufei Wei, Xiaoyong Li, Qing |
| contents | Large language models (LLMs), especially Explicit Long Chain-of-Thought (CoT) reasoning models like DeepSeek-R1 and QWQ, have demonstrated powerful reasoning capabilities, achieving impressive performance in commonsense reasoning and mathematical inference. Despite their effectiveness, Long-CoT reasoning models are often criticized for their limited ability and low efficiency in knowledge-intensive domains such as molecule discovery. Success in this field requires a precise understanding of domain knowledge, including molecular structures and chemical principles, which is challenging due to the inherent complexity of molecular data and the scarcity of high-quality expert annotations. To bridge this gap, we introduce Mol-R1, a novel framework designed to improve explainability and reasoning performance of R1-like Explicit Long-CoT reasoning LLMs in text-based molecule generation. Our approach begins with a high-quality reasoning dataset curated through Prior Regulation via In-context Distillation (PRID), a dedicated distillation strategy to effectively generate paired reasoning traces guided by prior regulations. Building upon this, we introduce MoIA, Molecular Iterative Adaptation, a sophisticated training strategy that iteratively combines Supervised Fine-tuning (SFT) with Reinforced Policy Optimization (RPO), tailored to boost the reasoning performance of R1-like reasoning models for molecule discovery. Finally, we examine the performance of Mol-R1 in the text-based molecule reasoning generation task, showing superior performance against existing baselines. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_08401 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Mol-R1: Towards Explicit Long-CoT Reasoning in Molecule Discovery Li, Jiatong Wang, Weida Zhang, Qinggang Li, Junxian Zhang, Di Zheng, Changmeng Zhang, Shufei Wei, Xiaoyong Li, Qing Computation and Language Large language models (LLMs), especially Explicit Long Chain-of-Thought (CoT) reasoning models like DeepSeek-R1 and QWQ, have demonstrated powerful reasoning capabilities, achieving impressive performance in commonsense reasoning and mathematical inference. Despite their effectiveness, Long-CoT reasoning models are often criticized for their limited ability and low efficiency in knowledge-intensive domains such as molecule discovery. Success in this field requires a precise understanding of domain knowledge, including molecular structures and chemical principles, which is challenging due to the inherent complexity of molecular data and the scarcity of high-quality expert annotations. To bridge this gap, we introduce Mol-R1, a novel framework designed to improve explainability and reasoning performance of R1-like Explicit Long-CoT reasoning LLMs in text-based molecule generation. Our approach begins with a high-quality reasoning dataset curated through Prior Regulation via In-context Distillation (PRID), a dedicated distillation strategy to effectively generate paired reasoning traces guided by prior regulations. Building upon this, we introduce MoIA, Molecular Iterative Adaptation, a sophisticated training strategy that iteratively combines Supervised Fine-tuning (SFT) with Reinforced Policy Optimization (RPO), tailored to boost the reasoning performance of R1-like reasoning models for molecule discovery. Finally, we examine the performance of Mol-R1 in the text-based molecule reasoning generation task, showing superior performance against existing baselines. |
| title | Mol-R1: Towards Explicit Long-CoT Reasoning in Molecule Discovery |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2508.08401 |