Latest Articles for Ethics In AI

A deep dive into ChatGPT's capabilities and constraints in natural language processing.

T-Explainer offers reliable insights into machine learning model predictions.

Soft Preference Optimization improves how language models align with human preferences.

This article presents a new method for enhancing reasoning in language models.

This study analyzes the effectiveness of LLMs in evaluating AI-generated explanations.

This study examines bias in Continual Learning models for medical imaging.

Examining the risks and opportunities of open-source generative AI technology.

A new method enhances AI responses by learning from multiple feedback sources.

MuDreamer enhances agent learning by focusing on prediction rather than image reconstruction.

Introducing Alignment from Demonstrations for safe and effective language models.

An overview of the growing field of text generation and its implications.

A new method blends weak and strong AI models to align with human values.

A study on how language models express and measure their confidence.

Explore how large language models enhance creativity through multimedia generation.

A look at how language models may represent beliefs and their implications.

This article explores the role of AI in comedy creation and the challenges faced by comedians.

Exploring the benefits of rule sets for ethical predictive modeling.

Setokim enhances the fusion of visual and text understanding through innovative tokenization.

A method to assess the reliability of AI outputs in healthcare and beyond.

A look at advancements in language models for Indic languages and their challenges.

Researchers develop GECO dataset and GECOBench to tackle gender bias in AI.

Explores the challenges of supervising advanced AI models with weaker counterparts.

New techniques aim to fix errors in language models without complete retraining.

A method to refine language models by reducing unwanted outputs during training.

Examining changes in social biases in language models over time.

Mamba’s context-extension method improves handling of long sequences without additional training.

Ensuring AI explanations are clear and reliable for user trust.

Research highlights safety neurons' role in enhancing LLM safety and responsibility.

Enhancing user engagement in large vision-language models through proactive communication.

Examining the hurdles LLMs face in low-resource language translation.

Dysca introduces a new way to assess LVLM performance using synthetic data.

Examining the challenges and implications of unlearning in AI models.

This article discusses how personas can improve synthetic data creation for training models.

Examining the relationship between privacy techniques and biases in language models.

Exploring how synthetic data shapes machine learning models and their behavior.

Recent research delves into the cognitive abilities of language models compared to humans.

Focus on Accessible Explainable AI for individuals with disabilities.

This study examines how LLMs evaluate their own knowledge and risk of errors.

Improving user experience by merging cognitive ergonomics with language models.

The Rashomon Effect reveals multiple effective models in machine learning.