Papers by Vishal Dey
\mathtt{GeLLM^3O}: Generalizing Large Language Models for Multi-property Molecule Optimization (2025.acl-long)
Copied to clipboard
| Challenge: | Large Language Models (LLMs) have remarkable out-of-domain generalizability to novel optimization tasks. |
| Approach: | They propose a series of instruction-tuned LLMs for molecule optimization that outperform state-of-the-art instruction-based LLM models. |
| Outcome: | mathttMuMOInstruct outperforms state-of-the-art LLMs on 5 in-domain and 5 out-of domain tasks. |
AutoSDT: Scaling Data-Driven Discovery Tasks Toward Open Co-Scientists (2025.emnlp-main)
Copied to clipboard
Yifei Li, Hanane Nour Moussa, Ziru Chen, Shijie Chen, Botao Yu, Mingyi Xue, Benjamin Burns, Tzu-Yao Chiu, Vishal Dey, Zitong Lu, Chen Wei, Qianheng Zhang, Tianyu Zhang, Song Gao, Xuhui Huang, Xia Ning, Nesreen K. Ahmed, Ali Payani, Huan Sun
| Challenge: | AutoSDT-5K is the only automatically collected and the largest open dataset for data-driven scientific discovery. |
| Approach: | They propose an automatic pipeline that collects high-quality coding tasks in real-world data-driven discovery workflows. |
| Outcome: | The proposed pipeline synthesizes accurate tasks and tasks from a dataset of 5,404 tasks covering four scientific disciplines and 756 Python packages. |
Large Language Models for Controllable Multi-property Multi-objective Molecule Optimization (2025.findings-emnlp)
Copied to clipboard
| Challenge: | Existing methods for molecule optimization fail to capture property-specific objectives . a series of instruction-tuned LLMs can perform targeted property-specific optimization . |
| Approach: | They propose a set of instruction-tuned LLMs that can perform targeted property-specific optimization. |
| Outcome: | a new instruction-tuned LLM can perform targeted property-specific optimization. |