Assessing foundation models’ abilities for human-level tasks is crucial for Artificial General Intelligence (AGI) development.Traditional benchmarks, which rely on artificial datasets, may not accurately represent these capabilities. In this paper, we introduce AGIEval, a novel bilingual benchmark designed to assess foundation models in the context of human-centric standardized exams, such as college entrance exams, law school admission tests, math competitions, and lawyer qualification tests. We evaluate several state-of-the-art foundation models on our benchmark. Impressively, we show that GPT-4 exceeds the average human performance in SAT, LSAT, and math contests, with 95% accuracy on SAT Math and 92.5% on the Chinese college entrance English exam. This demonstrates the exceptional performance of contemporary foundation models. In contrast, we also find that GPT-4 is less proficient in tasks requiring complex reasoning or specific domain knowledge. Our comprehensive analyses of model capabilities (understanding, knowledge, reasoning, and calculation) reveal their strengths and limitations, providing valuable insights into future directions for enhancing general capabilities. By concentrating on tasks pertinent to human cognition and decision-making, our benchmark delivers a meaningful and robust evaluation of foundation models’ performance in real-world scenarios.

Large language models (LLMs) have demonstrated impressive reasoning capabilities, yet there is ongoing debate about these abilities and the potential data contamination problem recently. This paper aims to evaluate the reasoning capacities of LLMs, specifically in solving recent competition-level programming problems in Codeforces, which are expert-crafted and unique, requiring deep understanding and robust reasoning skills. We first provide a comprehensive evaluation of GPT-4’s perceived zero-shot performance on this task, considering various aspects such as problems’ release time, difficulties, and types of errors encountered. Surprisingly, the perceived performance of GPT-4 has experienced a cliff like decline in problems after September 2021 consistently across all the difficulties and types of problems, which shows the potential data contamination, as well as the challenges for any existing LLM to solve unseen complex reasoning problems. We further explore various approaches such as fine-tuning, Chain-of-Thought prompting and problem description simplification. Unfortunately, none of them is able to consistently mitigate the challenges. Through our work, we emphasize the importance of this excellent data source for assessing the genuine reasoning capabilities of LLMs, and foster the development of LLMs with stronger reasoning abilities and better generalization in the future.

Large language models (LLMs) have exhibited remarkable ability in code generation. However, generating the correct solution in a single attempt still remains a challenge. Prior works utilize verification properties in software engineering to verify and re-rank solutions in a majority voting manner. But the assumption behind them that generated verification properties have better qualities than solutions may not always hold. In this paper, we treat them equally as different perspectives of LLMs’ reasoning processes. We propose the Multi-Perspective Self-Consistency (MPSC) framework incorporating both inter- and intra-consistency across outputs from multiple perspectives. Specifically, we prompt LLMs to generate diverse outputs from three perspectives, Solution, Specification and Test case, constructing a 3-partite graph. With two measure functions of consistency, we embed both inter- and intra-consistency information into the graph. The optimal choice of solutions is then determined based on analysis in the graph.MPSC significantly boosts performance of foundation models (ChatGPT in this paper) on various benchmarks, including HumanEval (+15.91%), MBPP (+6.43%) and CodeContests (+9.37%), even surpassing GPT-4.

Ensuring correctness is a pivotal aspect of software engineering. Among the various strategies available, software verification offers a definitive assurance of correctness. Nevertheless, writing verification proofs is resource-intensive and manpower-consuming, and there is a great need to automate this process. We introduce Selene in this paper, which is the first project-level automated proof benchmark constructed based on the real-world industrial-level operating system microkernel, seL4. Selene provides a comprehensive framework for end-to-end proof generation and a lightweight verification environment. Our experimental results with advanced large language models (LLMs), such as GPT-3.5-turbo and GPT-4, highlight the capabilities of LLMs in the domain of automated proof generation. Additionally, our further proposed augmentations indicate that the challenges presented by Selene can be mitigated in future research endeavors.

Code search is a task to find programming codes that semantically match the given natural language queries. Even though some of the existing datasets for this task are multilingual on the programming language side, their query data are only in English. In this research, we create a multilingual code search dataset in four natural and four programming languages using a neural machine translation model. Using our dataset, we pre-train and fine-tune the Transformer-based models and then evaluate them on multiple code search test sets. Our results show that the model pre-trained with all natural and programming language data has performed best in most cases. By applying back-translation data filtering to our dataset, we demonstrate that the translation quality affects the model’s performance to a certain extent, but the data size matters more.

Code execution is a fundamental aspect of programming language semantics that reflects the exact behavior of the code. However, most pre-trained models for code intelligence ignore the execution trace and only rely on source code and syntactic structures. In this paper, we investigate how well pre-trained models can understand and perform code execution. We develop a mutation-based data augmentation technique to create a large-scale and realistic Python dataset and task for code execution, which challenges existing models such as Codex. We then present CodeExecutor, a Transformer model that leverages code execution pre-training and curriculum learning to enhance its semantic comprehension. We evaluate CodeExecutor on code execution and show its promising performance and limitations. We also demonstrate its potential benefits for code intelligence tasks such as zero-shot code-to-code search and text-to-code generation. Our analysis provides insights into the learning and generalization abilities of pre-trained models for code execution.

One of the fundamental goals in code search is to retrieve a functionally correct code for a given natural language query. As annotating for correctness requires executing test cases (i.e. obtaining execution feedback), existing code search training datasets approximate text-code co-occurrences as positive execution feedback. However, this approximation may misalign models’ retrieval decisions from ground-truth correctness. To address such limitation, we propose Code Intervention-based Reinforcement Learning (CIRL) that perturbs training code to result in misalignment (i.e. code intervention), then tests models’ decisions and corrects them with the execution feedback by reinforcement learning. The first technical contribution of CIRL is to induce the execution feedback from perturbation, without actual execution. Secondly, CIRL introduces structural perturbations using abstract syntax trees, going beyond simple lexical changes. Experimental results on various datasets demonstrate the effectiveness of CIRL compared to conventional approaches.

Code completion, which aims to predict the following code token(s) according to the code context, can improve the productivity of software development. Recent work has proved that statistical language modeling with transformers can greatly improve the performance in the code completion task via learning from large-scale source code datasets. However, current approaches focus only on code context within the file or project, i.e. internal context. Our distinction is utilizing ”external” context, inspired by human behaviors of copying from the related code snippets when writing code. Specifically, we propose a retrieval-augmented code completion framework, leveraging both lexical copying and referring to code with similar semantics by retrieval. We adopt a stage-wise training approach that combines a source code retriever and an auto-regressive language model for programming language. We evaluate our approach in the code completion task in Python and Java programming languages, achieving a state-of-the-art performance on CodeXGLUE benchmark.

Pre-trained models for programming languages have recently demonstrated great success on code intelligence. To support both code-related understanding and generation tasks, recent works attempt to pre-train unified encoder-decoder models. However, such encoder-decoder framework is sub-optimal for auto-regressive tasks, especially code completion that requires a decoder-only manner for efficient inference. In this paper, we present UniXcoder, a unified cross-modal pre-trained model for programming language. The model utilizes mask attention matrices with prefix adapters to control the behavior of the model and leverages cross-modal contents like AST and code comment to enhance code representation. To encode AST that is represented as a tree in parallel, we propose a one-to-one mapping method to transform AST in a sequence structure that retains all structural information from the tree. Furthermore, we propose to utilize multi-modal contents to learn representation of code fragment with contrastive learning, and then align representations among programming languages using a cross-modal generation task. We evaluate UniXcoder on five code-related tasks over nine datasets. To further evaluate the performance of code fragment representation, we also construct a dataset for a new task, called zero-shot code-to-code search. Results show that our model achieves state-of-the-art performance on most tasks and analysis reveals that comment and AST can both enhance UniXcoder.

We study compositional generalization, which aims to generalize on unseen combinations of seen structural elements, for code search. Unlike existing approaches of partially pursuing this goal, we study how to extract structural elements, which we name a template that directly targets compositional generalization. Thus we propose CTBERT, or Code Template BERT, representing codes using automatically extracted templates as building blocks. We empirically validate CTBERT on two public code search benchmarks, AdvTest and CSN. Further, we show that templates are complementary to data flow graphs in GraphCodeBERT, by enhancing structural context around variables.

Producing the embedding of a sentence in anunsupervised way is valuable to natural language matching and retrieval problems in practice. In this work, we conduct a thorough examination of pretrained model based unsupervised sentence embeddings. We study on fourpretrained models and conduct massive experiments on seven datasets regarding sentence semantics. We have three main findings. First, averaging all tokens is better than only using [CLS] vector. Second, combining both topand bottom layers is better than only using toplayers. Lastly, an easy whitening-based vector normalization strategy with less than 10 linesof code consistently boosts the performance. The whole project including codes and data is publicly available at https://github.com/Jun-jie-Huang/WhiteningBERT.

Statistical language modeling and translation with transformers have found many successful applications in program understanding and generation tasks, setting high benchmarks for tools in modern software development environments. The finite context window of these neural models means, however, that they will be unable to leverage the entire relevant context of large files and packages for any given task. While there are many efforts to extend the context window, we introduce an architecture-independent approach for leveraging the syntactic hierarchies of source code for incorporating entire file-level context into a fixed-length window. Using concrete syntax trees of each source file we extract syntactic hierarchies and integrate them into context window by selectively removing from view more specific, less relevant scopes for a given task. We evaluate this approach on code generation tasks and joint translation of natural language and source code in Python programming language, achieving a new state-of-the-art in code completion and summarization for Python in the CodeXGLUE benchmark. We also introduce new CodeXGLUE benchmarks for user-experience-motivated tasks: code completion with normalized literals, method body completion/code summarization conditioned on file-level context.