A5100320726- Topic Modeling
- Machine Learning in Materials Science
- Computational Drug Discovery Methods
- Advanced Graph Neural Networks
- Natural Language Processing Techniques
- Artificial Intelligence in Law
- Gaze Tracking and Assistive Technology
- Visual Attention and Saliency Detection
- Law, Economics, and Judicial Systems
- Advanced Computing and Algorithms
- Diamond and Carbon-based Materials Research
- Academic Writing and Publishing
- Legal Education and Practice Innovations
- Conflict of Laws and Jurisdiction
- Data Quality and Management
- Carbon Nanotubes in Composites
- Thin-Film Transistor Technologies
National University of Singapore
2023-2024
Anhui University of Finance and Economics
2023
Xi'an Jiaotong University
2019
Northwest University
1990
Language Models (LMs) have demonstrated impressive molecule understanding ability on various 1D text-related tasks. However, they inherently lack 2D graph perception — a critical of human professionals in comprehending molecules’ topological structures. To bridge this gap, we propose MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter. MolCA enables an LM (i.e., Galactica) to understand both text- graph-based molecular contents via the cross-modal...
Language Models (LMs) have greatly influenced diverse domains. However, their inherent limitation in comprehending 3D molecular structures has considerably constrained potential the biomolecular domain. To bridge this gap, we focus on molecule-text interpretation, and propose 3D-MoLM: 3D-Molecular Modeling. Specifically, 3D-MoLM enables an LM to interpret analyze molecules by equipping with a encoder. This integration is achieved projector, bridging encoder's representation space LM's input...
In graph classification, the out-of-distribution (OOD) issue is attracting great attention. To address this issue, a prevailing idea to learn stable features, on assumption that they are substructures causally determining label and their relationship with distributional uncertainty. contrast, complementary parts termed environmental fail determine solely hold varying relationships label, thus ascribed possible reason for distribution shift. Existing generalization efforts mainly encourage...
Natural language is expected to be a key medium for various human-machine interactions in the era of large models. When it comes biochemistry field, series tasks around molecules (e.g., property prediction, molecule mining, etc.) are great significance while having high technical threshold. Bridging expressions natural and chemical can not only hugely improve interpretability reduce operation difficulty these tasks, but also fuse knowledge scattered complementary materials deeper...
Predicting chemical reactions, a fundamental challenge in chemistry, involves forecasting the resulting products from given reaction process. Conventional techniques, notably those employing Graph Neural Networks (GNNs), are often limited by insufficient training data and their inability to utilize textual information, undermining applicability real-world applications. In this work, we propose ReLM, novel framework that leverages knowledge encoded language models (LMs) assist GNNs, thereby...
Predicting chemical reactions, a fundamental challenge in chemistry, involves forecasting the resulting products from given reaction process. Conventional techniques, notably those employing Graph Neural Networks (GNNs), are often limited by insufficient training data and their inability to utilize textual information, undermining applicability real-world applications. In this work, we propose **ReLM**, novel framework that leverages knowledge encoded language models (LMs) assist GNNs,...
Partnership has a long history and lasting vitality. The earliest partnership is the product of capitalism result modern commercial trade development. With development human society, continues to improve improve, with advantages simple operation, high credit, flexible forms, etc. Today, become an extremely important form economic organization in market economy, widely existing free industries such as lawyers accountants. Because partnership, it used, so that gradually complicated, disputes...
Language Models (LMs) have demonstrated impressive molecule understanding ability on various 1D text-related tasks. However, they inherently lack 2D graph perception - a critical of human professionals in comprehending molecules' topological structures. To bridge this gap, we propose MolCA: Molecular Graph-Language Modeling with Cross-Modal Projector and Uni-Modal Adapter. MolCA enables an LM (e.g., Galactica) to understand both text- graph-based molecular contents via the cross-modal...