latest computer science research papers

Help | Advanced Search

Artificial Intelligence

Authors and titles for recent submissions.

• Fri, 15 Nov 2024
• Thu, 14 Nov 2024
• Wed, 13 Nov 2024
• Tue, 12 Nov 2024
• Mon, 11 Nov 2024

See today's new changes

Fri, 15 Nov 2024 (showing first 50 of 106 entries )

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.

• View all journals
• Explore content
• About the journal
• Publish with us
• Sign up for alerts

Computer science articles within Nature Computational Science

News Feature | 24 September 2024

The lost data: how AI systems censor LGBTQ+ content in the name of safety

Many AI companies implement safety systems to protect users from offensive or inaccurate content. Though well intentioned, these filters can exacerbate existing inequalities, and data shows that they have disproportionately removed LGBTQ+ content.

• Sophia Chen

Research Briefing | 19 July 2024

A multi-task learning strategy to pretrain models for medical image analysis

Pretraining powerful deep learning models requires large, comprehensive training datasets, which are often unavailable for medical imaging. In response, the universal biomedical pretrained (UMedPT) foundational model was developed based on multiple small and medium-sized datasets. This model reduced the amount of data required to learn new target tasks by at least 50%.

Article 19 July 2024 | Open Access

Overcoming data scarcity in biomedical imaging with a foundational multi-task model

UMedPT, a foundational model for biomedical imaging, has been trained on a variety of medical tasks with different types of label. It has achieved high performance with less training data in various clinical applications.

• Raphael Schäfer
• , Till Nicke
•  &  Fabian Kiessling

Article | 19 February 2024

Automated discovery of algorithms from data

Automated algorithm discovery has been difficult for artificial intelligence given the immense search space of possible functions. Here explainable neural networks are used to discover algorithms that outperform those designed by humans.

• Paul J. Blazek
• , Kesavan Venkatesh
•  &  Milo M. Lin

Correspondence | 21 December 2023

Using proprietary language models in academic research requires explicit justification

• Alexis Palmer
• , Noah A. Smith
•  &  Arthur Spirling

Research Highlight | 20 December 2023

One algorithm to play them all

• Fernando Chirigati

Research Briefing | 18 December 2023

A transformer method that predicts human lives from sequences of life events

Transformer methods are revolutionizing how computers process human language. Exploiting the structural similarity between human lives, seen as sequences of events, and natural-language sentences, a transformer method — dubbed life2vec — has been used to create rich vector representations of human lives, from which accurate predictions can be made.

Review Article | 21 November 2023

Designing molecules with autoencoder networks

Autoencoders are versatile tools for molecular informatics with the opportunity for advancing molecule and drug design. In this Review, the authors highlight the active areas of development in the field and explore the challenges that need to be addressed moving forward.

• Agnieszka Ilnicka
•  &  Gisbert Schneider

Research Highlight | 15 November 2023

A full-stack platform for spiking deep learning

Memory and computation together at last.

Comment | 26 October 2023

Building open-source AI

Artificial intelligence (AI) drives innovation across society, economies and science. We argue for the importance of building AI technology according to open-source principles to foster accessibility, collaboration, responsibility and interoperability.

• Yash Raj Shrestha
• , Georg von Krogh
•  &  Stefan Feuerriegel

Editorial | 10 October 2023

Ada Lovelace, a role model for the ages

Ada Lovelace Day celebrates women in STEM careers, but also raises awareness of the challenges that women have faced in science, as well as the importance of female role models in STEM.

Q&A | 10 October 2023

Laying the foundations of programming and system design

Dr Barbara Liskov — a mostly retired Institute Professor at the Massachusetts Institute of Technology, a pioneer in object-oriented programming and distributed systems and the winner of the 2008 ACM A. M. Turing Award, which is the highest distinction in computer science — talks to Nature Computational Science about her work on data abstractions, her career trajectory and recognizing the contributions of women in computer science.

• Ananya Rastogi

Brief Communication 05 October 2023 | Open Access

Human-like intuitive behavior and reasoning biases emerged in large language models but disappeared in ChatGPT

The reasoning capabilities of OpenAI’s generative pre-trained transformer family were tested using semantic illusions and cognitive reflection tests that are typically used in human studies. While early models were prone to human-like cognitive errors, ChatGPT decisively outperformed humans, avoiding the cognitive traps embedded in the tasks.

• Thilo Hagendorff
• , Sarah Fabi
•  &  Michal Kosinski

Research Highlight | 17 August 2023

Optimal crystal structure solutions

Perspective | 24 July 2023

Designing equitable algorithms

While the adherence to fairness constraints has become common practice in the design of algorithms across many contexts, a more holistic approach should be taken to avoid inflicting additional burdens on individuals in all groups, including those in marginalized communities.

• Alex Chohlas-Wood
• , Madison Coots
•  &  Julian Nyarko

Research Highlight | 20 April 2023

Moving toward safer driverless vehicles

Research Highlight | 23 January 2023

Large language model for molecular chemistry

News & Views | 05 January 2023

Dimensionality reduction under scrutiny

A framework for generating and interpreting dynamic visualizations from traditional static dimensionality reduction visualization methods has been proposed in a recent study.

• , Zewen K. Tuong
•  &  Di Yu

Resource | 30 December 2022

Dynamic visualization of high-dimensional data

Data visualization is widely used in science, but interpreting such visualizations is prone to error. Here a dynamic visualization is introduced for capturing more information and improving the reliability of visual interpretations.

• Eric D. Sun
•  &  James Zou

Research Highlight | 08 December 2022

Artificial agents that negotiate and reach agreements

• Iryna Omelchenko

Article | 28 November 2022

Combining computational controls with natural text reveals aspects of meaning composition

A neural network-based language model of supra-word meaning, that is, the combined meaning of words in a sentence, is proposed. Analysis of functional magnetic resonance imaging and magnetoencephalography data helps identify the regions of the brain responsible for understanding this meaning.

• Mariya Toneva
• , Tom M. Mitchell
•  &  Leila Wehbe

Editorial | 18 July 2022

Mathematics, the queen of sciences

We highlight how this year’s awardees from some of the most important prizes in mathematics have had an impact in the computational science community.

Editorial | 23 May 2022

Cracking the code review process

What does it entail to perform a code review for Nature Computational Science ?

Q&A | 01 May 2022

Fighting hate speech and misinformation online

Dr Srijan Kumar, assistant professor at Georgia Institute of Technology and a Forbes 30 Under 30 honoree in science, discusses with Nature Computational Science how he uses machine learning and data science to identify and mitigate malicious activities on online platforms, including misinformation and anti-Asian hate speech.

Research Highlight | 21 April 2022

Gauging urban development with neural networks

Editorial | 21 April 2022

And the Turing Award goes to…

The 2021 A. M. Turing Award celebrates Jack Dongarra’s contributions in high-performance computing, which have had a significant impact in computational science.

Q&A | 21 April 2022

A key player in high-performance computing

Dr Jack Dongarra, Distinguished Professor at the University of Tennessee and recipient of the 2021 A. M. Turing Award, spoke with Nature Computational Science about his contributions to high-performance computing (HPC) and his insights into the future of this field.

Q&A | 01 February 2022

Crypto and technology for the people

Dr Seny Kamara, associate professor at Brown University, talks to Nature Computational Science about his current research focus on the intersection between social responsibility and cryptography/technology, as well as about the need for multidisciplinary work in this arena.

Perspective | 31 January 2022

Opportunities for neuromorphic computing algorithms and applications

There is still a wide variety of challenges that restrict the rapid growth of neuromorphic algorithmic and application development. Addressing these challenges is essential for the research community to be able to effectively use neuromorphic computers in the future.

• Catherine D. Schuman
• , Shruti R. Kulkarni
•  &  Bill Kay

Resource 27 January 2022 | Open Access

The fast continuous wavelet transformation (fCWT) for real-time, high-quality, noise-resistant time–frequency analysis

The authors present an open-source framework that enables fast and accurate time–frequency analysis of signals and demonstrate it on real-world applications, such as signals from the brain–computer interface.

• Lukas P. A. Arts
•  &  Egon. L. van den Broek

Research Highlight | 21 January 2022

Machine learning to guide mathematicians

News & Views | 25 November 2021

Advancing data compression via noise detection

Compressing scientific data is essential to save on storage space, but doing so effectively while ensuring that the conclusions from the data are not affected remains a challenging task. A recent paper proposes a new method to identify numerical noise from floating-point atmospheric data, which can lead to a more effective compression.

• Dorit M. Hammerling
•  &  Allison H. Baker

Article 25 November 2021 | Open Access

Compressing atmospheric data into its real information content

Climate data are often stored at higher precision than is needed. The proposed compression automatically determines the precision from the data’s bitwise real information, removing any false information and leading to a more efficient compression.

• Milan Klöwer
• , Miha Razinger
•  &  Tim N. Palmer

Research Highlight | 12 November 2021

Accuracy and fairness go hand in hand

Accurate short-term precipitation prediction.

Correspondence | 11 October 2021

Democratizing interactive, immersive experiences for science education with WebXR

• Fabio Cortés Rodríguez
• , Matteo Dal Peraro
•  &  Luciano A. Abriata

Article | 22 September 2021

Explainable neural networks that simulate reasoning

The authors demonstrate how neural systems can encode cognitive functions, and use the proposed model to train robust, scalable deep neural networks that are explainable and capable of symbolic reasoning and domain generalization.

Research Highlight | 15 July 2021

Detection of war destruction from satellite images

Article | 24 June 2021

The power of quantum neural networks

A class of quantum neural networks is presented that outperforms comparable classical feedforward networks. They achieve a higher capacity in terms of effective dimension and at the same time train faster, suggesting a quantum advantage.

• Amira Abbas
• , David Sutter
•  &  Stefan Woerner

Research Highlight | 09 June 2021

Accurate and efficient fluid flow analysis

Comment | 13 May 2021

New views of black holes from computational imaging

The unique challenges associated with imaging a black hole motivated the development of new computational imaging algorithms. As the Event Horizon Telescope continues to expand, these algorithms will need to evolve to keep pace with the increasingly demanding volume and dimensionality of the data.

• Kazunori Akiyama
• , Andrew Chael
•  &  Dominic W. Pesce

News & Views | 25 March 2021

Efficient deep learning

The computational complexity of deep neural networks is a major obstacle of many application scenarios driven by low-power devices, including federated learning. A recent finding shows that random sketches can substantially reduce the model complexity without affecting prediction accuracy.

• Shiqiang Wang

Article | 25 March 2021

Random sketch learning for deep neural networks in edge computing

Developing lightweight deep neural networks, while essential for edge computing, still remains a challenge. Random sketch learning is a method that creates computationally efficient and compact networks, thus paving the way for deploying tiny machine learning (TinyML) in resource-constrained devices.

• , Peijun Chen
•  &  Jun Zhang

Research Highlight | 17 March 2021

Porting software without affecting scientific results

Article | 01 February 2021

Larger GPU-accelerated brain simulations with procedural connectivity

Spiking neural network simulations are very memory-intensive, limiting large-scale brain simulations to high-performance computer systems. Knight and Nowotny propose using procedural connectivity to substantially reduce the memory footprint of these models, such that they can run on standard GPUs.

• James C. Knight
•  &  Thomas Nowotny

Editorial | 14 January 2021

Celebrating today, inspiring tomorrow

Computational and mathematical models are at the core of myriad research developments across different domains. We pay tribute to the importance of computational science by providing a dedicated home among the Nature portfolio for this inspiring field.

Perspective | 14 January 2021

Quantifying causality in data science with quasi-experiments

While estimating causality from observational data is challenging, quasi-experiments provide causal inference methods with plausible assumptions that can be practical to a range of real-world problems.

• , Lyle Ungar
•  &  Konrad Kording

Browse broader subjects

• Systems biology
• Mathematics and computing

Quick links

• Explore articles by subject
• Guide to authors
• Editorial policies

Subscribe to the PwC Newsletter

Join the community, latest research, the denotational semantics of ssa.

imbrem/debruijn-ssa • 14 Nov 2024

Static single assignment form, or SSA, has been the dominant compiler intermediate representation for decades.

Programming Languages Logic in Computer Science F.3.2; D.3.4; D.3.1

PIMCOMP: An End-to-End DNN Compiler for Processing-In-Memory Accelerators

sunxt99/pimcomp-nn • 14 Nov 2024

How to deploy DNNs onto PIM-based accelerators is the key to explore PIM's high performance and energy efficiency.

Hardware Architecture

In Serverless, OS Scheduler Choice Costs Money: A Hybrid Scheduling Approach for Cheaper FaaS

ZhaoNeil/hybrid-scheduler • 13 Nov 2024

We present evidence that relying on the default Linux CFS scheduler increases serverless workloads cost by up to 10X.

Distributed, Parallel, and Cluster Computing

CorrectBench: Automatic Testbench Generation with Functional Self-Correction using LLMs for HDL Design

autobench/correctbench • 13 Nov 2024

The comparative analysis demonstrates that our method achieves a pass ratio of 70. 13% across all evaluated tasks, compared with the previous LLM-based testbench generation framework's 52. 18% and a direct LLM-based generation method's 33. 33%.

Software Engineering

Learning-Based Control Barrier Function with Provably Safe Guarantees: Reducing Conservatism with Heading-Aware Safety Margin

bassamlab/sigmarl • 13 Nov 2024

We propose a learning-based Control Barrier Function (CBF) to reduce conservatism in collision avoidance of car-like robots.

Robotics Multiagent Systems Systems and Control Systems and Control

RINO: Accurate, Robust Radar-Inertial Odometry with Non-Iterative Estimation

yangsc4063/rino • 12 Nov 2024

Additionally, the approach implements a loosely coupled system between the scanning radar and an inertial measurement unit (IMU), leveraging Error-State Kalman Filtering (ESKF).

Formalization of physics index notation in Lean 4

HEPLean/HepLean • 12 Nov 2024

The physics community relies on index notation to effectively manipulate types of tensors.

Logic in Computer Science High Energy Physics - Phenomenology High Energy Physics - Theory

A Framework for Carbon-aware Real-Time Workload Management in Clouds using Renewables-driven Cores

tharindu-b-hewage/openstack-gc • 12 Nov 2024

To this end, we present a framework to harvest green renewable energy for real-time workloads in cloud systems.

SoliDiffy: AST Differencing for Solidity Smart Contracts

mojtaba-eshghie/SoliDiffy • 12 Nov 2024

Smart contracts, primarily written in Solidity, are integral to blockchain software applications, yet precise analysis and maintenance are hindered by the limitations of existing differencing tools.

Software Engineering Programming Languages

Web-Based Simulator of Superscalar RISC-V Processors

Sekky61/riscv-sim • 12 Nov 2024

Mastering computational architectures is essential for developing fast and power-efficient programs.