HCI.TOPHCI, made easy

arrow_backBack to questions View categoryarrow_forward

General AI Systems and Interaction Design / Model Steering, Latent Space Editing, and Knowledge Injection

Edge devices have limited resources, making complex AI model training difficult; cloud training has privacy and latency issues.

lightbulbPractical problemModel Steering, Latent Space Editing, and Knowledge Injection

Edge devices have limited resources, making complex AI model training difficult; cloud training has privacy and latency issues.

UbiComp '21Context-Aware Compilation of DNN Training Pipelines across Edge and Cloud

Similar questions

lightbulbPractical problemModel Steering, Latent Space Editing, and Knowledge Injection

Users with epilepsy may face health threats from dynamic visual elements in mobile apps.

CHI '25MoWa: An Authoring Tool for Refining AI-Generated Human Avatar Motions Through Latent Waveform Manipulation

lightbulbPractical problemModel Steering, Latent Space Editing, and Knowledge Injection

Designers struggle to precisely control semantic attributes in text-to-image generation.

CHI '25AdaptiveSliders: User-aligned Semantic Slider-based Editing of Text-to-Image Model Output

lightbulbPractical problemModel Steering, Latent Space Editing, and Knowledge Injection

Non-expert users struggle to understand and improve deep learning models with existing methods.

IUI '24SpaceEditing: A Latent Space Editing Interface for Integrating Human Knowledge into Deep Neural Networks

lightbulbPractical problemModel Steering, Latent Space Editing, and Knowledge Injection

Users struggle to quickly find diverse images matching vague needs in exploratory image search.

IUI '24Sample, Nudge and Rank: Exploiting Interpretable GAN Controls for Exploratory Search

lightbulbPractical problemModel Steering, Latent Space Editing, and Knowledge Injection

In autonomous driving, low-visibility environments reduce accuracy of existing 3D detection methods.

UbiComp '243D Bounding Box Estimation Based on COTS mmWave Radar via Moving Scanning

lightbulbPractical problemModel Steering, Latent Space Editing, and Knowledge Injection

Intelligent applications on mobile devices struggle to guarantee performance under low-resource conditions.

UbiComp '23Genie in the Model: Automatic Generation of Human-in-the-Loop Deep Neural Networks for Mobile Applications

Related papers

MoWa: An Authoring Tool for Refining AI-Generated Human Avatar Motions Through Latent Waveform Manipulation

Jeongseok Oh, SeungJun Kim

AdaptiveSliders: User-aligned Semantic Slider-based Editing of Text-to-Image Model Output

Rahul Jain, Amit Goel, Koichiro Niinuma

SpaceEditing: A Latent Space Editing Interface for Integrating Human Knowledge into Deep Neural Networks

Jiafu Wei, Ding Xia, Haoran Xie

Sample, Nudge and Rank: Exploiting Interpretable GAN Controls for Exploratory Search

Yang Liu, Alan Medlar, Dorota Głowacka

Genie in the Model: Automatic Generation of Human-in-the-Loop Deep Neural Networks for Mobile Applications

Yanfei Wang, Zhiwen Yu, Sicong Liu

How Can Deep Neural Networks Aid Visualization Perception Research?: Three Studies on Correlation Judgments in Scatterplots

Fumeng Yang, Yuxin Ma, Lane Harrison

Continuous and Gradual Style Changes of Graphic Designs with Generative Model

Michihiko Ueno, Shin'ichi Satoh

DeepTake: Prediction of Driver Takeover Behavior using Multimodal Data

Erfan Pakdamanian, Shili Sheng, Sonia Baee