Publications

You can also find my articles on my Google Scholar profile.

ReactXT: Understanding Molecular” Reaction-ship” via Reaction-Contextualized Molecule-Text Pretraining

Published in findings of ACL, 2024

In this paper, we propose a novel pretraining paradigm, named ReactXT, to learn molecule-text representations that are contextualized by chemicalreaction contexts. We also introduce a new benchmark dataset, OpenExp, for evaluating the models on the task of experimental procedure prediction.

Download Paper

Rethinking Tokenizer and Decoder in Masked Graph Modeling for Molecules

Published in NeurIPS, 2024

In this paper, we summarize previous works on Masked Graph Modeling (MGM) for molecules and propose a novel MGM model, SimSGT, featuring a Simple GNN-based Tokenizer (SGT) and an effective decoding strategy.

Download Paper

ReLM: Leveraging Language Models for Enhanced Chemical Reaction Prediction

Published in findings of EMNLP, 2023

In this work, we propose ReLM, a novel framework that leverages the chemical knowledge encoded in language models (LMs) to assist GNNs, thereby enhancing the accuracy of real-world chemical reaction predictions.

Download Paper

Yaorui Shi

Publications

ReactXT: Understanding Molecular” Reaction-ship” via Reaction-Contextualized Molecule-Text Pretraining

Rethinking Tokenizer and Decoder in Masked Graph Modeling for Molecules

ReLM: Leveraging Language Models for Enhanced Chemical Reaction Prediction