helpResearch questionLLM Prompt Engineering and Authoring Tools
How can LLMs' instruction alignment performance be evaluated when executing complex multi-instruction prompts?Direction: LLM and Natural Language Interaction
LLM Prompt Engineering and Authoring Tools
Stats are based on currently indexed question data; missing sources remain visible.
40
items
10
sources
2025
latest
All questions
40 items
helpResearch questionLLM Prompt Engineering and Authoring Tools
Can user-involved dynamic evaluation frameworks improve prompt engineering efficiency and result quality?helpResearch questionLLM Prompt Engineering and Authoring Tools
Can decomposing and automatically generating evaluation criteria more precisely identify LLM alignment issues?lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
Users cannot efficiently evaluate and optimize LLM instruction effectiveness in complex tasks.helpResearch questionLLM Prompt Engineering and Authoring Tools
How can modular prompts be designed to optimize LLM performance in emotion and mental health state classification tasks?helpResearch questionLLM Prompt Engineering and Authoring Tools
Which persona and task instruction configurations in prompt design significantly affect model performance?helpResearch questionLLM Prompt Engineering and Authoring Tools
What impact do component interactions in modular prompt design have on task outcomes?lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
Developers struggle to optimize prompt design, leading to low accuracy in emotion and mental health tasks.helpResearch questionLLM Prompt Engineering and Authoring Tools
How can deeply integrating LLM-driven dialogue engines with GUIs create unified conversational user interfaces to improve data management experience?helpResearch questionLLM Prompt Engineering and Authoring Tools
How can semantic bottlenecks ensure structured output and improve interaction reliability and explainability in generated content?helpResearch questionLLM Prompt Engineering and Authoring Tools
How do user feedback and in-situ prompt engineering optimize dialogue control in semantic automation interfaces?lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
Data engineers and business analysts face inconsistent interfaces when integrating and processing complex data.helpResearch questionLLM Prompt Engineering and Authoring Tools
How can users' natural language feedback on LLMs be converted into persistent behavioral principles?helpResearch questionLLM Prompt Engineering and Authoring Tools
How can users effectively generate clear and specific principles through interactive evaluation to improve LLM performance?helpResearch questionLLM Prompt Engineering and Authoring Tools
How does ConstitutionMaker support users in identifying and converting improvement opportunities in model outputs?lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
Users' feedback on LLMs is difficult to sustain in model behavior, requiring repeated actions across interactions.helpResearch questionLLM Prompt Engineering and Authoring Tools
How accurate and relevant are prompts and guidance generated by LLM assistants in software help contexts?IUI '24Why and When LLM-Based Assistants Fail: Exploring the Effectiveness of Prompt-Based Interaction for Software Help-Seeking
helpResearch questionLLM Prompt Engineering and Authoring Tools
Can prompt generation integrating domain-specific information improve completion efficiency for complex software tasks?IUI '24Why and When LLM-Based Assistants Fail: Exploring the Effectiveness of Prompt-Based Interaction for Software Help-Seeking
helpResearch questionLLM Prompt Engineering and Authoring Tools
Do users form accurate mental models when interacting with LLMs?IUI '24Why and When LLM-Based Assistants Fail: Exploring the Effectiveness of Prompt-Based Interaction for Software Help-Seeking
lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
Lay users struggle to obtain accurate and easy-to-use help guidance when using complex software.IUI '24Why and When LLM-Based Assistants Fail: Exploring the Effectiveness of Prompt-Based Interaction for Software Help-Seeking
helpResearch questionLLM Prompt Engineering and Authoring Tools
How can users design and optimize LLM prompt chains from scratch?UIST '24ChainBuddy: An AI-assisted Agent System for Helping Users Set up LLM Pipelines
helpResearch questionLLM Prompt Engineering and Authoring Tools
How can AI assistant systems improve prompt chain setup efficiency and accuracy through task decomposition and multi-agent techniques?UIST '24ChainBuddy: An AI-assisted Agent System for Helping Users Set up LLM Pipelines
helpResearch questionLLM Prompt Engineering and Authoring Tools
What additional user support can interactive prompt chain optimization systems provide compared with existing tools?UIST '24ChainBuddy: An AI-assisted Agent System for Helping Users Set up LLM Pipelines
lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
Users struggle to design and optimize LLM prompt chains from scratch because the process is complex and laborious.UIST '24ChainBuddy: An AI-assisted Agent System for Helping Users Set up LLM Pipelines
helpResearch questionLLM Prompt Engineering and Authoring Tools
What specific challenges do non-AI experts face when designing LLM prompts?helpResearch questionLLM Prompt Engineering and Authoring Tools
Without programming background, how can ordinary users iteratively optimize prompts through tools?helpResearch questionLLM Prompt Engineering and Authoring Tools
What are non-expert users' thinking patterns and behavioral preferences in prompt design?lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
Ordinary users struggle to design effective prompts and optimize LLM output.helpResearch questionLLM Prompt Engineering and Authoring Tools
Can decomposing complex tasks through chaining improve transparency and controllability of large language models (LLMs)?helpResearch questionLLM Prompt Engineering and Authoring Tools
How can users effectively modify task chains through interactive interfaces to optimize task completion quality?helpResearch questionLLM Prompt Engineering and Authoring Tools
Which primitive operations most effectively decompose complex tasks and improve LLM generation quality?lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
Users struggle to understand and control large language model outputs, especially in multi-step complex tasks.helpResearch questionLLM Prompt Engineering and Authoring Tools
How do non-human metaphors affect user experience of conversational agents (e.g., engagement, cognitive load, intrinsic motivation, and trust)?helpResearch questionLLM Prompt Engineering and Authoring Tools
How do different task types (e.g., information retrieval and image classification) interact with metaphor design to affect user experience?helpResearch questionLLM Prompt Engineering and Authoring Tools
Can the "chain of being" framework provide effective theoretical tools for design beyond human-metaphor centrism?lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
When interacting with conversational agents, existing designs focus on human metaphors and overlook other possibilities.helpResearch questionLLM Prompt Engineering and Authoring Tools
Which prompt structures and hyperparameters significantly improve text-to-image model output quality?helpResearch questionLLM Prompt Engineering and Authoring Tools
How do subject-style interactions in text-to-image models affect final results?helpResearch questionLLM Prompt Engineering and Authoring Tools
What design strategies can reduce user trial-and-error costs when optimizing prompt structure?lightbulbPractical problemLLM Prompt Engineering and Authoring Tools
Users require extensive trial and error when using text-to-image models, with high learning costs.Related papers
IUI 2025
CoPrompter: User-Centric Evaluation of LM Instruction Alignment for Improved Prompt Engineering
Ishika Joshi, Simra Shahid, Shreeya Manasvi Venneti
CHI 2025
Exploring Modular Prompt Design for Emotion and Mental Health Recognition
Minseo Kim, Taemin Kim, Thu Hoang Anh Vo
IUI 2024
Empirical Evidence on Conversational Control of GUI in Semantic Automation
Daniel Karl I. Weidele, Mauro Martino, Abel N. Valente
IUI 2024
ConstitutionMaker: Interactively Critiquing Large Language Models by Converting Feedback into Principles
Savvas Petridis, Benjamin D Wedin, James Wexler
CHI 2023
Why Johnny Can’t Prompt: How Non-AI Experts Try (and Fail) to Design LLM Prompts
J.D. Zamfirescu-Pereira, Richmond Y. Wong, Bjoern Hartmann
CHI 2022
AI Chains: Transparent and Controllable Human-AI Interaction by Chaining Large Language Model Prompts
Tongshuang Wu, Michael Terry, Carrie J Cai
CHI 2022
Great Chain of Agents: The Role of Metaphorical Representation of Agents in Conversational Crowdsourcing
Ji-Youn Jung, Sihang Qiu, Alessandro Bozzon
CHI 2022
Design Guidelines for Prompt Engineering Text-to-Image Generative Models
Vivian Liu, Lydia B Chilton
Adjacent categories
LLM and Natural Language Interaction
LLM Interfaces, Prompts, and Interaction Understanding
141 items
LLM and Natural Language Interaction
Voice Assistant General Design and User Experience
94 items
LLM and Natural Language Interaction
Voice Interface Discoverability and Affordance Design
82 items
LLM and Natural Language Interaction
LLM User Dissatisfaction, Strategy Adjustment, and Satisfaction
80 items
LLM and Natural Language Interaction
Smart Speakers, Smart Homes, and Proactive Behavior Design
52 items
LLM and Natural Language Interaction
Voice Persona, Voice Quality, Prosody, and Social Trait Design
48 items
LLM and Natural Language Interaction
Information Organization, Document Analysis, and Qualitative Research
44 items
LLM and Natural Language Interaction
VUI Design Methods, Guidelines, and Heuristic Evaluation
44 items