CoLadder: Manipulating Code Generation via Multi-Level Blocks
Authors
This paper adopted an iterative design process to gain insights into programmers' strategies when using LLMs for programming. We proposed CoLadder, a novel system that supports programmers by facilitating hierarchical task decomposition, direct code segment manipulation, and result evaluation during prompt authoring. A user study with 12 experienced programmers showed that CoLadder is effective in helping programmers externalize their problem-solving intentions flexibly, improving their ability to evaluate and modify code across various abstraction levels, from their task's goal to final code implementation.
Research Questions / Practical Problems
Question signals indexed for this paper.
- 100%
Rapsai: Accelerating Machine Learning Prototyping of Multimedia Applications through Visual Programming
CHI '23· Human-LLM Collaboration +1
- 83%
Beyond Code Generation: LLM-supported Exploration of the Program Design Space
CHI '25· Generative AI (Text, Image, Music, Video) +2
- 83%
Linting Style and Substance in READMEs
CHI '26· Computational Methods in HCI +2
- 83%
The Way We Notice, That’s What Really Matters: Instantiating UI Components with Distinguishing Variations
CHI '26· Human-LLM Collaboration +2
- 83%
Athena: Intermediate Representations for Iterative Scaffolded App Generation with an LLM
IUI '26· Human-LLM Collaboration +2
- 80%
Understanding and Supporting Knowledge Decomposition for Machine Teaching
DIS '20· Human-LLM Collaboration +1
- 80%
Facilitating Knowledge Sharing from Domain Experts to Data Scientists for Building NLP Models
IUI '21· Human-LLM Collaboration +1
- 80%
Mallard: Turn the Web into a Contextualized Prototyping Environment for Machine Learning
UIST '19· Human-LLM Collaboration +1
- 80%
reCode: A Lightweight Find-and-Replace Interaction in the IDE for Transforming Code by Example
UIST '21· Human-LLM Collaboration +1
- 71%
Steering Semantic Data Processing With DocWrangler
UIST '25· Human-LLM Collaboration +2
Based on Jaccard similarity of research subtopics & professions (≥60%)