Albert Ziegler

Matteo Paltenghi, Rahul Pandita, Austin Z. Henley et al.

Recent neural models of code, such as OpenAI Codex and AlphaCode, have demonstrated remarkable proficiency at code generation due to the underlying attention mechanism. However, it often remains unclear how the models actually process code, and to what extent their reasoning and the way their attention mechanism scans the code matches the patterns of developers. A poor understanding of the model reasoning process limits the way in which current neural models are leveraged today, so far mostly for their raw prediction. To fill this gap, this work studies how the processed attention signal of three open large language models - CodeGen, InCoder and GPT-J - agrees with how developers look at and explore code when each answers the same sensemaking questions about code. Furthermore, we contribute an open-source eye-tracking dataset comprising 92 manually-labeled sessions from 25 developers engaged in sensemaking tasks. We empirically evaluate five heuristics that do not use the attention and ten attention-based post-processing approaches of the attention signal of CodeGen against our ground truth of developers exploring code, including the novel concept of follow-up attention which exhibits the highest agreement between model and human attention. Our follow-up attention method can predict the next line a developer will look at with 47% accuracy. This outperforms the baseline prediction accuracy of 42.3%, which uses the session history of other developers to recommend the next line. These results demonstrate the potential of leveraging the attention signal of pre-trained models for effective code exploration.

Albert Ziegler, Eirini Kalliamvakou, Shawn Simister et al.

Neural code synthesis has reached a point where snippet generation is accurate enough to be considered for integration into human software development workflows. Commercial products aim to increase programmers' productivity, without being able to measure it directly. In this case study, we asked users of GitHub Copilot about its impact on their productivity, and sought to find a reflection of their perception in directly measurable user data. We find that the rate with which shown suggestions are accepted, rather than more specific metrics regarding the persistence of completions in the code over time, drives developers' perception of productivity.

Albert Ziegler, Paweł Czyż · eth-zurich

Understanding how different classes are distributed in an unlabeled data set is an important challenge for the calibration of probabilistic classifiers and uncertainty quantification. Approaches like adjusted classify and count, black-box shift estimators, and invariant ratio estimators use an auxiliary (and potentially biased) black-box classifier trained on a different (shifted) data set to estimate the class distribution and yield asymptotic guarantees under weak assumptions. We demonstrate that all these algorithms are closely related to the inference in a particular Bayesian model, approximating the assumed ground-truth generative process. Then, we discuss an efficient Markov Chain Monte Carlo sampling scheme for the introduced model and show an asymptotic consistency guarantee in the large-data limit. We compare the introduced model against the established point estimators in a variety of scenarios, and show it is competitive, and in some cases superior, with the state of the art.

Tobias Kiecker, Jan Arne Sparka, Martin Reuter et al.

Maintaining consistency between code and documentation is a crucial yet frequently overlooked aspect of software development. Even minor mismatches can confuse API users, introduce new bugs, and increase overall maintenance effort. This creates demand for automated solutions that can assist developers in identifying code-documentation inconsistencies. However, since automatic reports still require human confirmation, false positives carry serious consequences: wasting developer time and discouraging practical adoption. We introduce CASCADE (Consistency Analysis for Source Code And Documentation through Execution), a novel tool for detecting inconsistencies with a strong emphasis on reducing false positives. CASCADE leverages Large Language Models (LLMs) to generate unit tests directly from natural-language documentation. Since these tests are derived from the documentation, any failure during execution indicates a potential mismatch between the documented and actual behavior of the code. To minimize false positives, CASCADE also generates code from the documentation to cross-check the generated tests. By design, an inconsistency is reported only when two conditions are met: the existing code fails a test, while the code generated from the documentation passes the same test. We evaluated CASCADE on a novel dataset of 71 inconsistent and 814 consistent code-documentation pairs drawn from open-source Java projects. Further, we applied CASCADE to additional Java, C#, and Rust repositories, where we uncovered 13 previously unknown inconsistencies, of which 10 have subsequently been fixed, demonstrating both CASCADE's precision and its applicability to real-world codebases.

Albert Ziegler, Paweł Czyż

Unsupervised recalibration (URC) is a general way to improve the accuracy of an already trained probabilistic classification or regression model upon encountering new data while deployed in the field. URC does not require any ground truth associated with the new field data. URC merely observes the model's predictions and recognizes when the training set is not representative of field data, and then corrects to remove any introduced bias. URC can be particularly useful when applied separately to different subpopulations observed in the field that were not considered as features when training the machine learning model. This makes it possible to exploit subpopulation information without retraining the model or even having ground truth for some or all subpopulations available. Additionally, if these subpopulations are the object of study, URC serves to determine the correct ground truth distributions for them, where naive aggregation methods, like averaging the model's predictions, systematically underestimate their differences.

Ian Wright, Albert Ziegler

We apply machine learning to version control data to measure the quantity of effort required to produce source code changes. We construct a model of a `standard coder' trained from examples of code changes produced by actual software developers together with the labor time they supplied. The effort of a code change is then defined as the labor hours supplied by the standard coder to produce that change. We therefore reduce heterogeneous, structured code changes to a scalar measure of effort derived from large quantities of empirical data on the coding behavior of software developers. The standard coder replaces traditional metrics, such as lines-of-code or function point analysis, and yields new insights into what code changes require more or less effort.