Latest Articles for Machine Learning

Explore the principles and applications of stochastic approximation in uncertain environments.

A study on improving LLMs' problem-solving abilities using a new framework.

GloVe-V improves word embeddings by measuring uncertainty in vector representations.

Investigating how latent space affects transformer model performance on language tasks.

This article analyzes how language models understand and infer causal relationships.

A fresh approach to analyze particle collisions for new physics insights.

Methods for identifying important features in low-quality data environments.

New methods enhance understanding of brain activity through neural data analysis.

A new method combines human and AI for efficient text classification.

A new framework enhances 3D scene representation using limited 2D images.

This article explores the impact of data poisoning on language model alignment.

Research explores connections between top quark, Higgs boson, and CP violation.

A new benchmark for evaluating dynamic text-attributed graphs in various fields.

A new model increases accuracy in natural language understanding by using expert predictions.

APTP improves text-to-image models for better efficiency and quality.

This article discusses soft prompting as a method for machine unlearning in LLMs.

Self-MoE creates specialized experts for improved language model performance.

New techniques enhance the efficiency of solving large linear systems.

Research explores drag reduction in cylinder flow with active control methods and machine learning.

Leveraging language models improves predictions for tabular data across various fields.

P3GNN enhances APT detection while protecting data privacy in SDN networks.

New method enhances conversational effectiveness in language models through planning techniques.

Learn how transcoders help clarify complex language models.

Examining the limitations of large language models in understanding code relationships.

RepoExec evaluates code generation performance at the repository level.

A framework improves code generation for specialized languages using documentation.

A new approach to enhance image-text matching using dual-encoder models and scene graphs.

An analysis of how LLMs learn and retain factual information.

A new dataset improves multi-document reasoning for eligibility questions.

Research shows simple sensors can rival complex cameras in robot tasks.

This paper presents an efficient approach for real-time change detection in data.

A method to learn low-dimensional dynamics from noisy high-dimensional observations.

SCEPTR offers a new way to predict TCR specificity using sparse data efficiently.

This article examines generative models that can outperform human experts in chess.

A new framework helps language models learn symbolic language without human input.

Exploring how quantum computing can enhance machine learning techniques.

A new tool uses machine learning to detect performance bugs effectively.

Examining memorization in code completion models and its privacy implications.

A new model that speeds up high-quality image creation using innovative techniques.

The Nemotron-4 340B family delivers powerful models for diverse applications and synthetic data generation.