createLiveAI LogocreateLiveAI
Page 111 of 130 • 1560 Total Articles

createLiveAI

Continue exploring the latest AI breakthroughs, technology insights, and industry analysis. Page 111 of our comprehensive AI news collection.

All Articles 1560 Business 249 Ethics 150 General 142 Policy 12 Research 793 Startups 13 Technology 201

📰 Latest Intelligence

Showing 12 articles on page 111 of 130

Live feed
arXiv Machine Learning
Research
Daunce: Data Attribution through Uncertainty Estimation - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

Daunce: Data Attribution through Uncertainty Estimation

The paper introduces Daunce, a scalable and accurate data attribution method that estimates influence by analyzing the covariance of losses across perturbed models, making it suitable for large models like LLMs and proprietary systems such as GPT. Unlike gradient-based approaches, Daunce leverages uncertainty estimation through model perturbations, enhancing attribution accuracy for applications like data debugging and curation.

research gpt +2
Read More
arXiv Machine Learning
Research
Decom-Renorm-Merge: Model Merging on the Right Space Improves Multitasking - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

Decom-Renorm-Merge: Model Merging on the Right Space Improves Multitasking

A new method called Decom-Renorm-Merge (DRM) has been proposed to improve neural network model merging by using Singular Value Decomposition to align weight matrices into a joint space, enabling more effective merging across various architectures. DRM outperforms existing techniques in both full finetuning and low-rank adaptation scenarios, with renormalization identified as a key factor for creating a robust merging process.

research machine-learning +1
Read More
arXiv Machine Learning
Research
DeepRTE: Pre-trained Attention-based Neural Network for Radiative Tranfer - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

DeepRTE: Pre-trained Attention-based Neural Network for Radiative Tranfer

Researchers introduced DeepRTE, a neural network method utilizing pre-trained attention mechanisms to accurately and efficiently solve the steady-state Radiative Transfer Equation, which models radiation propagation in various scientific fields. Numerical experiments demonstrate the approach's high accuracy and computational benefits across applications like atmospheric transfer, heat transfer, and optical imaging.

Deep Learning Transformers
Read More
arXiv Machine Learning
Research
Defining Foundation Models for Computational Science: A Call for Clarity and Rigor - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

Defining Foundation Models for Computational Science: A Call for Clarity and Rigor

This paper highlights the need for a clear, formal definition of foundation models in computational science, emphasizing core qualities like generality, reusability, and scalability. It introduces the Data-Driven Finite Element Method (DD-FEM), which combines traditional numerical methods with data-driven learning to address challenges such as scalability and physics consistency, providing a foundation for future development in the field.

Machine Learning Computer Vision +1
Read More
arXiv Machine Learning
Research
>- - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

>-

A new method called DenoiseRotator improves pruning of large language models by redistributing parameter importance through learnable orthogonal transformations, making models more robust to pruning under semi-structured sparsity. Evaluations on models like LLaMA3 and Qwen2.5 show significant reductions in performance degradation, with the approach enhancing perplexity and zero-shot accuracy compared to traditional pruning techniques.

research machine-learning
Read More
arXiv Machine Learning
Research
Development and Validation of SXI++ LNM Algorithm for Sepsis Prediction - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

Development and Validation of SXI++ LNM Algorithm for Sepsis Prediction

A new machine learning model, SXI++ LNM, utilizing deep neural networks, significantly improves sepsis prediction accuracy, achieving an AUC of 0.99, with a precision of 99.9% and an accuracy of 99.99%. The model demonstrates high robustness across various datasets, outperforming existing methods in clinical scenarios for early sepsis detection.

research machine-learning +1
Read More
arXiv Machine Learning
Research
DINGO: Constrained Inference for Diffusion LLMs - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

DINGO: Constrained Inference for Diffusion LLMs

Diffusion LLMs are a promising, efficient alternative to autoregressive models but struggle to enforce formal constraints like regular expressions, limiting their reliability for structured output tasks. To address this, the authors propose DINGO, a dynamic programming-based decoding method that efficiently and provably preserves the model's distribution while strictly satisfying user-defined constraints, significantly improving constrained generation performance.

research llm +1
Read More
arXiv Machine Learning
Research
Directed Graph Grammars for Sequence-based Learning - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

Directed Graph Grammars for Sequence-based Learning

This work introduces a grammar-based method to represent directed acyclic graphs (DAGs) as unique, sequential derivations over an unambiguous grammar, enabling lossless compression and principled decoding. This compact representation facilitates applications such as graph generation, property prediction, and Bayesian optimization by providing a continuous sequence-based structure for complex DAGs.

research machine-learning
Read More
arXiv Machine Learning
Research
Does Machine Unlearning Truly Remove Model Knowledge? A Framework for Auditing Unlearning in LLMs - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

Does Machine Unlearning Truly Remove Model Knowledge? A Framework for Auditing Unlearning in LLMs

This paper introduces a comprehensive auditing framework to evaluate the effectiveness of machine unlearning algorithms in removing sensitive information from Large Language Models (LLMs), addressing privacy and ownership concerns. It includes benchmark datasets, multiple unlearning methods, and novel techniques such as intermediate activation perturbations to improve robustness beyond traditional prompt-based assessments.

Transformers
Read More
arXiv Machine Learning
Research
$K^2$VAE: A Koopman-Kalman Enhanced Variational AutoEncoder for Probabilistic Time Series Forecasting - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

$K^2$VAE: A Koopman-Kalman Enhanced Variational AutoEncoder for Probabilistic Time Series Forecasting

The paper introduces $K^2$VAE, a VAE-based generative model that improves long-term probabilistic time series forecasting by transforming nonlinear dynamics into a linear system using KoopmanNet and refining predictions with KalmanNet, thereby reducing error accumulation. Extensive experiments show that $K^2$VAE outperforms existing methods in both short- and long-term forecasting, offering a more efficient and accurate approach.

Read More
arXiv Machine Learning
Research
DOPPLER: Dual-Policy Learning for Device Assignment in Asynchronous Dataflow Graphs - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

DOPPLER: Dual-Policy Learning for Device Assignment in Asynchronous Dataflow Graphs

The paper introduces Doppler, a three-stage framework employing dual-policy networks to optimize operation placement in dataflow graphs, reducing execution time for machine learning workloads. Unlike previous methods, Doppler addresses limitations of bulk-synchronous systems, system scheduling awareness, and reliance solely on reinforcement learning, leading to improved efficiency and training performance.

Machine Learning
Read More
arXiv Machine Learning
Research
Efficiently Access Diffusion Fisher: Within the Outer Product Span Space - AI news coverage from arXiv Machine Learning in Research
Research
📄 arXiv Machine Learning

Efficiently Access Diffusion Fisher: Within the Outer Product Span Space

This paper reveals that the diffusion Fisher information in diffusion models can be efficiently approximated using outer products of the score and initial data, rather than relying on computationally intensive auto-differentiation. The proposed algorithms offer improved accuracy and reduced computational cost, with applications demonstrated in likelihood evaluation, optimization, and verifying properties of diffusion maps.

Read More
Previous Page
1 2 3 4 5 6 7 ... 130
Next Page

Page 111 of 130 • Showing articles 1321-1332 of 1560

Quick Navigation

Jump to any page or browse by category

Latest (Page 1) Business 249 Ethics 150 General 142 Policy 12 Research 793 Startups 13