AttentionBender applies 2D transforms to cross-attention maps in video diffusion transformers, producing distributed distortions and glitch aesthetics that reveal entangled attention mechanisms while serving as both an XAI probe and creative tool.
hub Canonical reference
Emerging roles and relationships among humans and interactive ai systems.International Journal of Human–Computer Interaction, 41(17):10595–10617
Canonical reference. 73% of citing Pith papers cite this work as background.
15
Pith papers citing it
Background
73% of classified citations
hub tools
citation-role summary
background 10
method 1
citation-polarity summary
representative citing papers
Qualitative study of 19 practitioners reveals ten LLM product evaluation practices and introduces the results-actionability gap as a key barrier to turning findings into improvements.
EmoMM benchmark reveals Video Contribution Collapse in MLLMs for emotion recognition under modality conflict and missingness, mitigated by CHASE head-level attention steering.
LLM originality raters exhibit self-preference bias toward artificial responses that disappears after controlling for idea elaboration in the Alternate Uses Task.
Each tested LLM shows its own characteristic unreliability when engaging in repair during extended math-question dialogues.
Adaptive Prompt Elicitation (APE) uses an information-theoretic framework to generate visual queries that elicit and compile user intent into better prompts for text-to-image models, showing improved alignment in benchmarks and a user study.
13 participants became convinced AI understands human values after chatbot interactions evaluated with the VAPT toolkit.
Develops and validates a two-factor scale for machine companionship with AI companions via exploratory factor analysis on N=467 and confirmation on N=249, with post-hoc identification of socioinstrumental and autotelic templates.
Creatives prefer self-experimentation over structured guidance for GenAI image tools to preserve creative freedom, even when guidance aids AI literacy.
A co-located tablet VR setup with spatial separation and tool asymmetry enabled dynamic coordination and active use of teamwork KSAs in two collaborative scenarios.
AI-labeled input devices raise user performance expectations but produce no measurable change in objective or subjective interaction outcomes.
Hiding generative AI use to signal expertise reduces knowledge sharing and transparency among workplace colleagues.
UTAUT is suitable for studying individual barriers to GenAI use in software engineering when combined with Bayesian analysis, with three priorities for future research on construct refinement, operationalization, and statistical methods.
Researchers clustered 41,300 Moltbook posts from AI agents with k-means and retrieval-augmented generation to produce validated personas that represent behavioral diversity in agent populations.
citing papers explorer
-
AttentionBender: Manipulating Cross-Attention in Video Diffusion Transformers as a Creative Probe
AttentionBender applies 2D transforms to cross-attention maps in video diffusion transformers, producing distributed distortions and glitch aesthetics that reveal entangled attention mechanisms while serving as both an XAI probe and creative tool.
-
Results-Actionability Gap: Understanding How Practitioners Evaluate LLM Products in the Wild
Qualitative study of 19 practitioners reveals ten LLM product evaluation practices and introduces the results-actionability gap as a key barrier to turning findings into improvements.
-
EmoMM: Benchmarking and Steering MLLM for Multimodal Emotion Recognition under Conflict and Missingness
EmoMM benchmark reveals Video Contribution Collapse in MLLMs for emotion recognition under modality conflict and missingness, mitigated by CHASE head-level attention steering.
-
The Effect of Idea Elaboration on the Automatic Assessment of Idea Originality
LLM originality raters exhibit self-preference bias toward artificial responses that disappears after controlling for idea elaboration in the Alternate Uses Task.
-
Talking to a Know-It-All GPT or a Second-Guesser Claude? How Repair reveals unreliable Multi-Turn Behavior in LLMs
Each tested LLM shows its own characteristic unreliability when engaging in repair during extended math-question dialogues.
-
Adaptive Prompt Elicitation for Text-to-Image Generation
Adaptive Prompt Elicitation (APE) uses an information-theoretic framework to generate visual queries that elicit and compile user intent into better prompts for text-to-image models, showing improved alignment in benchmarks and a user study.
-
AI and My Values: User Perceptions of LLMs' Ability to Extract, Embody, and Explain Human Values from Casual Conversations
13 participants became convinced AI understands human values after chatbot interactions evaluated with the VAPT toolkit.
-
Measuring Machine Companionship: Scale Development and Validation for AI Companions
Develops and validates a two-factor scale for machine companionship with AI companions via exploratory factor analysis on N=467 and confirmation on N=249, with post-hoc identification of socioinstrumental and autotelic templates.
-
How Creatives Approach GenAI Image Generation: Tensions Between Structured Guidance, Self-Experimentation, and Creative Autonomy
Creatives prefer self-experimentation over structured guidance for GenAI image tools to preserve creative freedom, even when guidance aids AI literacy.
-
Where's the Team Spirit? An Exploratory Study on Team Development Through Co-located Tablet-Based VR
A co-located tablet VR setup with spatial separation and tool asymmetry enabled dynamic coordination and active use of teamwork KSAs in two collaborative scenarios.
-
AI Washing Inflates Expected Performance but Not Interaction Outcomes: An AI Placebo Study Using Fitts' Law
AI-labeled input devices raise user performance expectations but produce no measurable change in objective or subjective interaction outcomes.
-
"If You're Very Clever, No One Knows You've Used It": The Social Dynamics of Developing Generative AI Literacy in the Workplace
Hiding generative AI use to signal expertise reduces knowledge sharing and transparency among workplace colleagues.
-
GenAI in Software Engineering: The Role of Technology Acceptance Models
UTAUT is suitable for studying individual barriers to GenAI use in software engineering when combined with Bayesian analysis, with three priorities for future research on construct refinement, operationalization, and statistical methods.
-
How to Model AI Agents as Personas?: Applying the Persona Ecosystem Playground to 41,300 Posts on Moltbook for Behavioral Insights
Researchers clustered 41,300 Moltbook posts from AI agents with k-means and retrieval-augmented generation to produce validated personas that represent behavioral diversity in agent populations.
- LLM Harms: A Taxonomy and Discussion