Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain
the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in
Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles
and JavaScript.
Designing and optimizing proteins by mutagenesis suffers from the overwhelming space of possible variants. A recent study developed µProtein, a reinforcement learning model coupled with a protein language model as a surrogate oracle, to accelerate this process towards high-functioning proteins.
Ilievski et al. examine differences and similarities in the various ways human and AI systems generalize. The insights are important for effectively supporting alignment in human–AI teams.
The next major challenge for artificial intelligence in drug development lies in proving its value in real-world settings. A new technology not only supports the generation of novel chemical entities but also accelerates a range of real-world molecular design tasks.
Thoughtfully designed negative training datasets may hold the key to more robust machine learning models. Ursu et al. reveal how negative training data composition shapes antibody prediction models and their generalizability. Sometimes, the best way to get better is to train harder.
Despite impressive performances of current large AI models, symbolic and abstract reasoning tasks often elicit failure modes in these systems. In this Perspective, Ito et al. propose to make use of computational complexity theory, formulating algebraic problems as computable circuits to address the challenge of mathematical and symbolic reasoning in AI systems.
A unified optimization framework, CORNETO, introduces a versatile approach to knowledge-driven biological network inference, bringing machine learning sensibilities to systems biology.
By integrating multi-dimensional data with deep learning, a new method known as ImmuScope predicts both major histocompatibility complex class II (MHC-II) presentation and T helper cell immunogenicity. ImmuScope shows potential to accelerate neoantigen discovery and vaccine design.
Inherited retinal diseases are both numerous and diverse, but all arise from genetic mutations leading to retinal degeneration. Through the use of modern diagnostic tools, accurate genotype prediction is now possible using high-resolution imaging techniques alone, facilitating improved screening and genetic variant prioritization.
Predicting the macroscopic properties of molecular liquids from first principles is a major challenge owing to the disordered nature of liquids and the weak link between microscopic forces and thermodynamic observables. A new workflow called BAMBOO produces accurate and transferable machine learning interatomic potential simulations of liquid electrolytes.
Miret and Krishnan discuss the promise of large language models (LLMs) to revolutionize materials discovery via automated processing of complex, interconnected, multimodal materials data. They also consider critical limitations and research opportunities needed to unblock LLMs for breakthroughs in materials science.
Various machine learning models have been developed in recent years for the discovery of crystal structures. Matbench Discovery, a new benchmark, offers an efficient way to identify the most promising architectures.
Don-Yehiya et al. explore creating an open ecosystem for human feedback on large language models, drawing from peer-production, open-source and citizen-science practices, and addressing key challenges to establish sustainable feedback loops between users and specialized models.
AI technologies are advancing rapidly, offering new solutions for autonomous robot operation in complex environments. Aude Billard et al. discuss the need to identify and adapt AI technologies for robotics, proposing a research roadmap to address key challenges and opportunities.
While transformers and large language models excel at efficiently processing long sequences, new approaches have been proposed that incorporate recurrence to overcome the quadratic cost of self-attention. Tiezzi et al. discuss recurrent and state-space models and the promise they hold for future sequence processing networks.
AI tools are increasingly used for important decisions, but they can be uncertain about specific individuals or groups. Chakraborty et al. discuss the need for better methods to assess uncertainty in high-stakes applications such as healthcare and finance, and outline a set of main challenges to provide practical guidance for AI researchers.
A systematic review of peer-reviewed AI safety research reveals extensive work on practical and immediate concerns. The findings advocate for an inclusive approach to AI safety that embraces diverse motivations and perspectives.
More than 30 years have passed since the advent of omics technologies revolutionized biological and medical research. Research now highlights the unique opportunity to integrate and decode complex biological mechanisms for health and diseases with machine learning.
Micaela Consens et al. discuss and review the recent rise of transformer-based and large language models in genomics. They also highlight promising directions for genome language models beyond the transformer architecture.
