Uncovering Weaknesses in Neural Code Generation
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911957768994816 |
|---|---|
| author | Lian, Xiaoli Wang, Shuaisong Ma, Jieping Liu, Fang Tan, Xin Zhang, Li Shi, Lin Gao, Cuiyun |
| author_facet | Lian, Xiaoli Wang, Shuaisong Ma, Jieping Liu, Fang Tan, Xin Zhang, Li Shi, Lin Gao, Cuiyun |
| contents | Code generation, the task of producing source code from prompts, has seen significant advancements with the advent of pre-trained large language models (PLMs). Despite these achievements, there lacks a comprehensive taxonomy of weaknesses about the benchmark and the generated code, which risks the community's focus on known issues at the cost of under-explored areas.
Our systematic study aims to fill this gap by evaluating five state-of-the-art PLMs: three larger models, CodeGen2.5 with 7 billion parameters, CodeGeeX2 with 6 billion parameters, GPT-4 Turbo, and two smaller ones, UnixCoder with 110 million parameters and CodeT5 base with 220 million parameters, across three popular datasets, CoNaLa, HumanEval Plus, and DS-1000. We assess the quality of generated code using match-based and execution-based metrics, then conduct thematic analysis to develop a taxonomy of nine types of weaknesses.
We dissected weakness distributions in both larger and smaller models, applying an extensive methodology that encompasses model-specific as well as collective analysis (union and intersection) across models. Our research uncovers three salient findings: 1. In the CoNaLa dataset, inaccurate prompts are a notable problem, causing all large models to fail in 26.84% of cases, with even higher failure rates of 40% for smaller models; 2. Missing pivotal semantics is a pervasive issue across benchmarks, with one or more large models omitting key semantics in 65.78% of CoNaLa tasks, and similarly high occurrences in HumanEval Plus (66.09%) and DS-1000 (80.51%); 3. All models struggle with proper API usage, a challenge amplified by vague or complex prompts.
Our findings aim to steer researchers towards addressing specific weaknesses and challenges in code generation. Furthermore, our annotations can offer a targeted benchmark subset for detailed analysis. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_09793 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Uncovering Weaknesses in Neural Code Generation Lian, Xiaoli Wang, Shuaisong Ma, Jieping Liu, Fang Tan, Xin Zhang, Li Shi, Lin Gao, Cuiyun Software Engineering Code generation, the task of producing source code from prompts, has seen significant advancements with the advent of pre-trained large language models (PLMs). Despite these achievements, there lacks a comprehensive taxonomy of weaknesses about the benchmark and the generated code, which risks the community's focus on known issues at the cost of under-explored areas. Our systematic study aims to fill this gap by evaluating five state-of-the-art PLMs: three larger models, CodeGen2.5 with 7 billion parameters, CodeGeeX2 with 6 billion parameters, GPT-4 Turbo, and two smaller ones, UnixCoder with 110 million parameters and CodeT5 base with 220 million parameters, across three popular datasets, CoNaLa, HumanEval Plus, and DS-1000. We assess the quality of generated code using match-based and execution-based metrics, then conduct thematic analysis to develop a taxonomy of nine types of weaknesses. We dissected weakness distributions in both larger and smaller models, applying an extensive methodology that encompasses model-specific as well as collective analysis (union and intersection) across models. Our research uncovers three salient findings: 1. In the CoNaLa dataset, inaccurate prompts are a notable problem, causing all large models to fail in 26.84% of cases, with even higher failure rates of 40% for smaller models; 2. Missing pivotal semantics is a pervasive issue across benchmarks, with one or more large models omitting key semantics in 65.78% of CoNaLa tasks, and similarly high occurrences in HumanEval Plus (66.09%) and DS-1000 (80.51%); 3. All models struggle with proper API usage, a challenge amplified by vague or complex prompts. Our findings aim to steer researchers towards addressing specific weaknesses and challenges in code generation. Furthermore, our annotations can offer a targeted benchmark subset for detailed analysis. |
| title | Uncovering Weaknesses in Neural Code Generation |
| topic | Software Engineering |
| url | https://arxiv.org/abs/2407.09793 |