Unifying Molecular and Textual Representations via Multi-task Language Modelling

TL;DR

This paper introduces a multi-task language model that unifies chemical and natural language representations, enabling efficient cross-domain task solving without domain-specific fine-tuning, thus advancing data-driven scientific discovery.

Contribution

It presents the first multi-domain, multi-task language model capable of handling both chemical and natural language tasks simultaneously, improving performance through shared representations.

Findings

Sharing weights across domains improves task performance.

Model outperforms state-of-the-art baselines on multiple benchmarks.

Cross-domain sharing enhances results more as scale increases.

Abstract

The recent advances in neural language models have also been successfully applied to the field of chemistry, offering generative solutions for classical problems in molecular design and synthesis planning. These new methods have the potential to fuel a new era of data-driven automation in scientific discovery. However, specialized models are still typically required for each task, leading to the need for problem-specific fine-tuning and neglecting task interrelations. The main obstacle in this field is the lack of a unified representation between natural language and chemical representations, complicating and limiting human-machine interaction. Here, we propose the first multi-domain, multi-task language model that can solve a wide range of tasks in both the chemical and natural language domains. Our model can handle chemical and natural language concurrently, without requiring expensive pre-training on single domains or task-specific models. Interestingly, sharing weights across domains remarkably improves our model when benchmarked against state-of-the-art baselines on single-domain and cross-domain tasks. In particular, sharing information across domains and tasks gives rise to large improvements in cross-domain tasks, the magnitude of which increase with scale, as measured by more than a dozen of relevant metrics. Our work suggests that such models can robustly and efficiently accelerate discovery in physical sciences by superseding problem-specific fine-tuning and enhancing human-model interactions.