Latest Articles for Human-Computer Interaction

Exploring how AI improves dialogue-based understanding and response accuracy.

A new dataset generates real-time action suggestions for egocentric video viewers.

A dataset to test language models' grasp of wording differences.

Exploring annotated portfolios to improve robot designs in public spaces.

Centaurs combine human insight and AI to improve decision-making across various fields.

A clear framework to assess understanding in AI systems.

A method aims to enhance the accuracy of AI-generated responses.

This study assesses how well language models understand social situations using a game.

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

An exploration of AI's role in work and human interactions.

Error Span Annotation offers a fast and reliable approach to translation quality assessment.

This study examines methods to enhance machine empathy through storytelling.

LLMs assist human raters in effectively identifying harmful online content.

A study on how language models generate persuasive rationales for argument evaluation.

This benchmark evaluates multimodal models on BPM tasks like documentation and improvement.

Examining how LLMs exhibit personality traits through new testing methods.

Combining EEG and fTCD improves communication for individuals with disabilities.

A new method for enhancing machine learning through suboptimal examples and human feedback.

Introducing iWISDM, a new benchmark for assessing machine instruction following with visuals and language.

LiveMind enhances language models for faster, real-time interactions with users.

Introducing a method to minimize overoptimization in models trained with human feedback.

Study shows touch improves memory for virtual objects.

Current models struggle with spatial reasoning, relying more on text than images.

A new method to improve AI alignment with human values using corrupted feedback.

A study on how machines adapt to phonological changes in speech.

A comprehensive study on language models’ performance across 10 Indic languages.

A new system improves real-time hand gesture recognition for various applications.

GIEBench assesses LLMs' empathetic responses based on diverse group identities.

Research shows models struggle with step dependencies in cooking recipes.

A look at how AI models interpret human choices and the need for improvement.

Investigating fine-grained feedback for text-to-image models and its practical implications.

A new method to define rewards for reinforcement learning agents using language models.

Study assesses how well MLLMs interpret visual data and their performance compared to humans.

A new method improves reconstruction of hand-face interactions for AR and VR.

Can self-play enhance language models' performance in cooperative settings?

GeoHOI enhances human-object interaction detection using geometric features for improved accuracy.

A new approach to classify human and machine-generated texts more effectively.

Introducing a method to create realistic 3D scenes for virtual reality and design.

New model improves accuracy in detecting emotions during spoken exchanges.

New training methods enhance language models' ability to create detailed long texts.