Interpretable Program Synthesis

Zhang, Tianyi; Chen, Ziyang; Zhu, Yuanli; Vaithilingam, Priyan; Wang, Xinyu; Glassman, Elena L.

doi:10.1145/3411764.3445646

Cited by 15 publications

(4 citation statements)

References 51 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Code Generation. Code generation models [7,9,10] and program synthesis techniques [6,12,50,61] enable users to complete tasks without programming by using easier specifications, including natural language, examples, and demonstrations. Code generation models like Codex [7], PaLM [9], and InCoder [10] are transformer-based causal language models (commonly referred to as LLMs) that complete texts from natural language prompts.…”

Section: Related Workmentioning

confidence: 99%

“…Because code generation techniques generalize programs from incomplete user specifications, generated programs are inherently ambiguous, and thus require disambiguation to identify a correct solution among candidates. Prior work proposes techniques to visualize the search process [61], visualize code candidates [52,59], and present distinguishing examples for authors to inspect [15]. Data Formulator provides feedback to the authors by presenting the generated code together with its execution results for them to inspect, select, and edit.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Data Formulator: AI-powered Concept-driven Visualization Authoring

Wang,

Thompson,

Lee

2023

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

With most modern visualization tools, authors need to transform their data into tidy formats to create visualizations they want. Because this requires experience with programming or separate data processing tools, data transformation remains a barrier in visualization authoring. To address this challenge, we present a new visualization paradigm, concept binding, that separates high-level visualization intents and low-level data transformation steps, leveraging an AI agent. We realize this paradigm in Data Formulator, an interactive visualization authoring tool. With Data Formulator, authors first define data concepts they plan to visualize using natural languages or examples, and then bind them to visual channels. Data Formulator then dispatches its AI-agent to automatically transform the input data to surface these concepts and generate desired visualizations. When presenting the results (transformed table and output visualizations) from the AI agent, Data Formulator provides feedback to help authors inspect and understand them. A user study with 10 participants shows that participants could learn and use Data Formulator to create visualizations that involve challenging data transformations, and presents interesting future research directions.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Data Formulator: AI-powered Concept-driven Visualization Authoring

Wang,

Thompson,

Lee

2023

IEEE Trans. Visual. Comput. Graphics

View full text Add to dashboard Cite

show abstract

“…For example, millions of Microsoft Excel users use FlashFill [27], a PbE string transformation system. Some of the above approaches, such as PbE, were developed around user needs and involved human-centered evaluation [59], [60], [75], [76]. In addition, many user studies have revealed several major usability issues and challenges in these synthesis systems [28], [39], [40], [57], which led to the design of better synthesis systems.…”

Section: Background and Related Workmentioning

confidence: 99%

Using Reactive Synthesis: An End-to-End Exploratory Case Study

Maoz

2023

2023 IEEE/ACM 45th International Conference on Software Engineering (ICSE)

View full text Add to dashboard Cite

“…In addition, code search spaces can become large, which can lead either to lengthy synthesis delays or too many suggestions. Concerns like these have led to work on interpretable synthesis [21] and interactive search space exploration [22].…”

Section: Introductionmentioning

confidence: 99%

An Integrative Human-Centered Architecture for Interactive Programming Assistants

Blinn

Moon

Griffis

et al. 2022

2022 IEEE Symposium on Visual Languages and Human-Centric Computing (VL/HCC)

View full text Add to dashboard Cite

Programming has become a collaboration between human programmers, who drive intent, and interactive assistants that suggest contextually relevant editor actions. There has been considerable work on suggestion synthesis strategies-from semantic autocomplete to modern program synthesis, repair, and machine learning research. This diversity of contextually viable strategies creates a need for an integrative, human-centered perspective on the problem of programming assistant design that (1) confronts the problem of integrating a variety of synthesis strategies, fed by shared semantic analyses capable of operating on program sketches, and (2) centers the needs of the human programmer: comprehending, comparing, ranking, and filtering suggestions generated by various synthesizers, and in some cases participating in a synthesizer's search by supplying additional expressions of intent. This paper contributes a conceptual architecture and API to guide programming assistant designers as they confront these integration and human-centered design challenges. We then instantiate this architecture with two prototype end-toend assistant designs, both developed for the Hazel programming environment, that emphasize understudied design aspects, namely continuity, explainability, human-in-the-loop synthesis, and the integration of multiple analyses with multiple synthesis strategies.

show abstract

Interpretable Program Synthesis

Cited by 15 publications

References 51 publications

Data Formulator: AI-powered Concept-driven Visualization Authoring

Data Formulator: AI-powered Concept-driven Visualization Authoring

Using Reactive Synthesis: An End-to-End Exploratory Case Study

An Integrative Human-Centered Architecture for Interactive Programming Assistants

Contact Info

Product

Resources

About