Yu Wang

Shuyang Jiang, Yuhao Wang, Yu Wang

Large language models (LLMs) have already revolutionized code generation, after being pretrained on publicly available code data. However, while various methods have been proposed to augment LLMs with retrieved knowledge and enhance the quality of code generation, the performance of these retrieval-based methods is limited by the strength of the retrievers used. In addition, while LLMs show great emergent ability, they still struggle to produce the correct code in one turn. To address these challenges, we propose a novel two-step pipeline, called \autoknow, that leverages LLMs as both knowledge providers and self-reflective programmers. Unlike retrieval-based methods, \autoknow~obtains the knowledge from input prompts and generates intermediate code based on the generated knowledge. After that, \autoknow~asks LLM to act as an expert programmer to perform debugging for the generated code. This is achieved by receiving the error message from the interpreter, without requiring special test cases for correctness verification. We evaluate \autoknow~on three code generation datasets, including DS-1000 for data science code, HumanEval for software engineering code, and TransCoder for C++-to-Python translation. Our empirical experiments show that \autoknow~outperforms strong baselines by a significant margin on all datasets. We also conduct exhaustive analytical experiments to validate the effectiveness of the two stages of \autoknow, and find that both are superior to other prompting-based methods. Further scalability analysis demonstrates that \autoknow~can be adapted to other more advanced models, such as GPT-4, and bring consistent efficacy improvement.

Chi Zhang, Yu Wang, Linzhang Wang

The recent breakthroughs in deep learning methods have sparked a wave of interest in learning-based bug detectors. Compared to the traditional static analysis tools, these bug detectors are directly learned from data, thus, easier to create. On the other hand, they are difficult to train, requiring a large amount of data which is not readily available. In this paper, we propose a new approach, called meta bug detection, which offers three crucial advantages over existing learning-based bug detectors: bug-type generic (i.e., capable of catching the types of bugs that are totally unobserved during training), self-explainable (i.e., capable of explaining its own prediction without any external interpretability methods) and sample efficient (i.e., requiring substantially less training data than standard bug detectors). Our extensive evaluation shows our meta bug detector (MBD) is effective in catching a variety of bugs including null pointer dereference, array index out-of-bound, file handle leak, and even data races in concurrent programs; in the process MBD also significantly outperforms several noteworthy baselines including Facebook Infer, a prominent static analysis tool, and FICS, the latest anomaly detection method.

Yulin Ou, Yu Wang, Yang Xu et al.

Recent theoretical advancement of information density in natural language has brought the following question on desk: To what degree does natural language exhibit periodicity pattern in its encoded information? We address this question by introducing a new method called AutoPeriod of Surprisal (APS). APS adopts a canonical periodicity detection algorithm and is able to identify any significant periods that exist in the surprisal sequence of a single document. By applying the algorithm to a set of corpora, we have obtained the following interesting results: Firstly, a considerable proportion of human language demonstrates a strong pattern of periodicity in information; Secondly, new periods that are outside the distributions of typical structural units in text (e.g., sentence boundaries, elementary discourse units, etc.) are found and further confirmed via harmonic regression modeling. We conclude that the periodicity of information in language is a joint outcome from both structured factors and other driving factors that take effect at longer distances. The advantages of our periodicity detection method and its potentials in LLM-generation detection are further discussed.

Yu Wang, Ke Wang, Linzhang Wang

Attribution methods have emerged as a popular approach to interpreting model predictions based on the relevance of input features. Although the feature importance ranking can provide insights of how models arrive at a prediction from a raw input, they do not give a clear-cut definition of the key features models use for the prediction. In this paper, we present a new method, called WheaCha, for explaining the predictions of code models. Although WheaCha employs the same mechanism of tracing model predictions back to the input features, it differs from all existing attribution methods in crucial ways. Specifically, WheaCha divides an input program into "wheat" (i.e., the defining features that are the reason for which models predict the label that they predict) and the rest "chaff" for any prediction of a learned code model. We realize WheaCha in a tool, HuoYan, and use it to explain four prominent code models: code2vec, seq-GNN, GGNN, and CodeBERT. Results show (1) HuoYan is efficient - taking on average under twenty seconds to compute the wheat for an input program in an end-to-end fashion (i.e., including model prediction time); (2) the wheat that all models use to predict input programs is made of simple syntactic or even lexical properties (i.e., identifier names); (3) Based on wheat, we present a novel approach to explaining the predictions of code models through the lens of training data.

Yu Wang, Fengjuan Gao, Linzhang Wang et al.

Learning distributed representations of source code has been a challenging task for machine learning models. Earlier works treated programs as text so that natural language methods can be readily applied. Unfortunately, such approaches do not capitalize on the rich structural information possessed by source code. Of late, Graph Neural Network (GNN) was proposed to learn embeddings of programs from their graph representations. Due to the homogeneous and expensive message-passing procedure, GNN can suffer from precision issues, especially when dealing with programs rendered into large graphs. In this paper, we present a new graph neural architecture, called Graph Interval Neural Network (GINN), to tackle the weaknesses of the existing GNN. Unlike the standard GNN, GINN generalizes from a curated graph representation obtained through an abstraction method designed to aid models to learn. In particular, GINN focuses exclusively on intervals for mining the feature representation of a program, furthermore, GINN operates on a hierarchy of intervals for scaling the learning to large graphs. We evaluate GINN for two popular downstream applications: variable misuse prediction and method name prediction. Results show in both cases GINN outperforms the state-of-the-art models by a comfortable margin. We have also created a neural bug detector based on GINN to catch null pointer deference bugs in Java code. While learning from the same 9,000 methods extracted from 64 projects, GINN-based bug detector significantly outperforms GNN-based bug detector on 13 unseen test projects. Next, we deploy our trained GINN-based bug detector and Facebook Infer to scan the codebase of 20 highly starred projects on GitHub. Through our manual inspection, we confirm 38 bugs out of 102 warnings raised by GINN-based bug detector compared to 34 bugs out of 129 warnings for Facebook Infer.

Yu Wang, Yining Sun, Zuchang Ma et al.

Named Entity Recognition (NER) is a fundamental Natural Language Processing (NLP) task to extract entities from unstructured data. The previous methods for NER were based on machine learning or deep learning. Recently, pre-training models have significantly improved performance on multiple NLP tasks. In this paper, firstly, we introduce the architecture and pre-training tasks of four common pre-training models: BERT, ERNIE, ERNIE2.0-tiny, and RoBERTa. Then, we apply these pre-training models to a NER task by fine-tuning, and compare the effects of the different model architecture and pre-training tasks on the NER task. The experiment results showed that RoBERTa achieved state-of-the-art results on the MSRA-2006 dataset.

Yu Wang, Fengjuan Gao, Linzhang Wang et al.

We present an alternative approach to creating static bug finders. Instead of relying on human expertise, we utilize deep neural networks to train static analyzers directly from data. In particular, we frame the problem of bug finding as a classification task and train a classifier to differentiate the buggy from non-buggy programs using Graph Neural Network (GNN). Crucially, we propose a novel interval-based propagation mechanism that leads to a significantly more efficient, accurate and scalable generalization of GNN. We have realized our approach into a framework, NeurSA, and extensively evaluated it. In a cross-project prediction task, three neural bug detectors we instantiate from NeurSA are effective in catching null pointer dereference, array index out of bound and class cast bugs in unseen code. We compare NeurSA against several static analyzers (e.g. Facebook Infer and Pinpoint) on a set of null pointer dereference bugs. Results show that NeurSA is more precise in catching the real bugs and suppressing the spurious warnings. We also apply NeurSA to several popular Java projects on GitHub and discover 50 new bugs, among which 9 have been fixed, and 3 have been confirmed.