Computer Science - Machine Learning

A new method for combining training data enhances language model performance.

Study explores how machine learning models predict CLABSI risk using hospital data.

Researchers use Brainscore to evaluate AI models against human brain activity.

Study reveals language models utilize complex, multi-dimensional features for language processing.

Using advanced sensors during training helps robots perform tasks better.

Robots combining wheels and legs aim to enhance urban logistics and deliveries.

PV-Tuning improves fine-tuning and compression for large language models.

A new tool for measuring cluster separation in data analysis.

Discover how tensor networks can transform various industrial applications and optimize processes.

Examining fairness and ethics of how language models respond to users.

A new method for understanding preferences using feature-level comparisons.

This article reviews the performance of language models in public health data analysis.

This research investigates using large language models for detecting anomalies in time series data.

This study explores methods for improving anomaly detection in procurement transactions.

This article discusses enhancing code completion tools by predicting developer needs for suggestions.

This paper examines how sensors enhance decisions despite incomplete data.

AGILE enhances LLM agents with memory, expert advice, and real-time tools.

Introducing AnyLoss, transforming metrics into loss functions for better model training.

A look into causal relationships and their importance in various fields.

Examining the link between local differential privacy and fairness in ML systems.

Discover how machine learning is advancing synthetic biology through innovative methods.

A look at measuring and tackling error accumulation in atmospheric modeling.

This article proposes a method for selecting games to determine rankings fairly.

A new method improves personalized image generation using existing AI models.

Federated Batch Normalization improves model training in distributed learning environments.

Dynamic datasets enhance model learning and reduce resource needs.

I2CL improves efficiency and performance in language model tasks.

PhiNet mimics human memory systems for improved machine learning.

TimeMixer combines detailed and broad data patterns for better forecasting accuracy.

ShapeFormer enhances classification accuracy by combining class-specific and general features.

This method merges deep learning and math for improved image inpainting.

A study introduces RepNoise to protect large language models from harmful fine-tuning.

A model improves predictions of soil's light reflection to aid environmental studies.

New methods enhance predictions and parameter estimation in physical simulations.

A new method boosts AI performance by effectively using all available experts.

A new method combines technology for accurate carbon footprint calculation and understandable explanations.

This article presents a method for identifying causal relationships in complex systems.

New methods improve privacy auditing, focusing on hidden state models in machine learning.

LoRA-Ensemble combines models for better predictions and uncertainty management.

A new system enhances machine learning through improved abstract reasoning and fewer examples.