David Cao
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
David Cao has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Max Willsey
University of California, Berkeley, Berkeley, USA
2 shared publications - Zachary Tatlock
University of Washington, Seattle, USA
2 shared publications - Chandrakana Nandi
Certora, Seattle, USA
1 shared publication - Eli Rosenthal
Google, Mountain View, USA
1 shared publication - Justin Lubin
University of California at Berkeley, Berkeley, USA
1 shared publication - Nadia Polikarpova
University of California at San Diego, La Jolla, USA
1 shared publication
Latest Publications
- Open Access
- 1 October 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Programming Languages
- Vol. 10 (OOPSLA2) , 1132-1163
Direct Manipulation and Natural Language Programming, Together at Last?
Decades of programming languages research has contributed novel approaches to program editing that go beyond modifying text, including direct manipulation programming, structure editing, and automated refactoring tools. However, the rapid growth of natural language programming largely reinforces a view of programs as text and program editing as (unstructured) text transformation. How can we develop unified programming systems that bridge the gap between these approaches, supporting multiple editing paradigms in concert? And how would such systems change the way we program? We take a first step toward answering these questions by introducing a framework that enables program editing via both direct manipulation and natural language, and instantiate this framework in a variant of the cartokit direct manipulation programming system ( cartokit DM+NL ). Our key insight is to treat programs as sequences of structured edits and to use an edit language as a shared interface for both direct manipulation and natural language interactions, leveraging constrained decoding to support the latter. Using our instantiation, we conducted a within-subjects study ( N =18) to understand how the combination of direct manipulation and natural language as editing modalities changes the programming process compared to each modality alone. Perhaps surprisingly, we found that study participants overwhelmingly chose to edit via direct manipulation when both modalities were available, performing just 6.14% of edits via natural language. Our thematic analysis of study sessions revealed that direct manipulation aided task decomposition, encouraged incremental editing, and helped mitigate known challenges in natural language programming related to understanding model capabilities and interpreting model-generated code. Conversely, natural language editing came into play largely to automate, parameterize, and replay known edits that would otherwise be repeated tediously by hand. Our edit-based framework and study findings lay out a possible pathway for future research on programming systems that blend natural language with alternative editing modalities, building on the foundation of edit languages.
- 1 October 2026
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Programming Languages
- Vol. 10 (OOPSLA2) , 1132-1163
Open AccessDirect Manipulation and Natural Language Programming, Together at Last?
Decades of programming languages research has contributed novel approaches to program editing that go beyond modifying text, including direct manipulation programming, structure editing, and automated refactoring tools. However, the rapid growth of natural language programming largely reinforces a view of programs as text and program editing as (unstructured) text transformation. How can we develop unified programming systems that bridge the gap between these approaches, supporting multiple editing paradigms in concert? And how would such systems change the way we program? We take a first step toward answering these questions by introducing a framework that enables program editing via both direct manipulation and natural language, and instantiate this framework in a variant of the cartokit direct manipulation programming system ( cartokit DM+NL ). Our key insight is to treat programs as sequences of structured edits and to use an edit language as a shared interface for both direct manipulation and natural language interactions, leveraging constrained decoding to support the latter. Using our instantiation, we conducted a within-subjects study ( N =18) to understand how the combination of direct manipulation and natural language as editing modalities changes the programming process compared to each modality alone. Perhaps surprisingly, we found that study participants overwhelmingly chose to edit via direct manipulation when both modalities were available, performing just 6.14% of edits via natural language. Our thematic analysis of study sessions revealed that direct manipulation aided task decomposition, encouraged incremental editing, and helped mitigate known challenges in natural language programming related to understanding model capabilities and interpreting model-generated code. Conversely, natural language editing came into play largely to automate, parameterize, and replay known edits that would otherwise be repeated tediously by hand. Our edit-based framework and study findings lay out a possible pathway for future research on programming systems that blend natural language with alternative editing modalities, building on the foundation of edit languages.
- Open Access
- 6 June 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Programming Languages
- Vol. 7 (PLDI) , 468-492
Better Together: Unifying Datalog and Equality Saturation
We present egglog, a fixpoint reasoning system that unifies Datalog and equality saturation (EqSat). Like Datalog, egglog supports efficient incremental execution, cooperating analyses, and lattice-based reasoning. Like EqSat, egglog supports term rewriting, efficient congruence closure, and extraction of optimized terms. We identify two recent applications -- a unification-based pointer analysis in Datalog and an EqSat-based floating-point term rewriter -- that have been hampered by features missing from Datalog but found in EqSat or vice-versa. We evaluate our system by reimplementing those projects in egglog. The resulting systems in egglog are faster, simpler, and fix bugs found in the original systems.
- 6 June 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Programming Languages
- Vol. 7 (PLDI) , 468-492
Open AccessBetter Together: Unifying Datalog and Equality Saturation
We present egglog, a fixpoint reasoning system that unifies Datalog and equality saturation (EqSat). Like Datalog, egglog supports efficient incremental execution, cooperating analyses, and lattice-based reasoning. Like EqSat, egglog supports term rewriting, efficient congruence closure, and extraction of optimized terms. We identify two recent applications -- a unification-based pointer analysis in Datalog and an EqSat-based floating-point term rewriter -- that have been hampered by features missing from Datalog but found in EqSat or vice-versa. We evaluate our system by reimplementing those projects in egglog. The resulting systems in egglog are faster, simpler, and fix bugs found in the original systems.
- Open Access
- 9 January 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Programming Languages
- Vol. 7 (POPL) , 396-424
babble: Learning Better Abstractions with E-Graphs and Anti-unification
Library learning compresses a given corpus of programs by extracting common structure from the corpus into reusable library functions. Prior work on library learning suffers from two limitations that prevent it from scaling to larger, more complex inputs. First, it explores too many candidate library functions that are not useful for compression. Second, it is not robust to syntactic variation in the input. We propose library learning modulo theory (LLMT), a new library learning algorithm that additionally takes as input an equational theory for a given problem domain. LLMT uses e-graphs and equality saturation to compactly represent the space of programs equivalent modulo the theory, and uses a novel e-graph anti-unification technique to find common patterns in the corpus more directly and efficiently. We implemented LLMT in a tool named babble. Our evaluation shows that babble achieves better compression orders of magnitude faster than the state of the art. We also provide a qualitative evaluation showing that babble learns reusable functions on inputs previously out of reach for library learning.
- 9 January 2023
- Published by Association for Computing Machinery (ACM) in Proceedings of the ACM on Programming Languages
- Vol. 7 (POPL) , 396-424
Open Accessbabble: Learning Better Abstractions with E-Graphs and Anti-unification
Library learning compresses a given corpus of programs by extracting common structure from the corpus into reusable library functions. Prior work on library learning suffers from two limitations that prevent it from scaling to larger, more complex inputs. First, it explores too many candidate library functions that are not useful for compression. Second, it is not robust to syntactic variation in the input. We propose library learning modulo theory (LLMT), a new library learning algorithm that additionally takes as input an equational theory for a given problem domain. LLMT uses e-graphs and equality saturation to compactly represent the space of programs equivalent modulo the theory, and uses a novel e-graph anti-unification technique to find common patterns in the corpus more directly and efficiently. We implemented LLMT in a tool named babble. Our evaluation shows that babble achieves better compression orders of magnitude faster than the state of the art. We also provide a qualitative evaluation showing that babble learns reusable functions on inputs previously out of reach for library learning.