An official website of the United States government
A significant amount of research is focused on developing and evaluating large language models for a variety of code synthesis tasks. These include synthesizing code from natural language, synthesizing tests from code, and synthesizing explanations of code. In contrast, the behavior of instructional code editing with LLMs is understudied. These are tasks in which the model is provided a block of code and an instruction to modify the code. The editing instruction may ask for a feature to be added or removed, describe a bug and ask for a fix, or ask for a different kind of solution. We introduce a carefully crafted benchmark of code editing tasks and use it to evaluate several cutting edge LLMs. Our evaluation exposes a significant gap between the capabilities of state-of-the-art open and closed models. For example, even GPT-3.5-Turbo is better than the best open model at code editing tasks. We also introduce a new, carefully curated, permissively licensed training dataset of code editing tasks coupled with natural language instructions. Using this training dataset, we show that we can fine-tune open Code LLMs to significantly improve their code editing capabilities, closing the gap between open and closed models. All code, data, and models are available at https://github.com/nuprl/CanItEdit.
more »
« less
Assessment of chemistry knowledge in large language models that generate code
In this work, we investigate the question: do code-generating large language models know chemistry? Our results indicate, mostly yes. To evaluate this, we introduce an expandable framework for evaluating chemistry knowledge in these models, through prompting models to solve chemistry problems posed as coding tasks. To do so, we produce a benchmark set of problems, and evaluate these models based on correctness of code by automated testing and evaluation by experts. We find that recent LLMs are able to write correct code across a variety of topics in chemistry and their accuracy can be increased by 30 percentage points via prompt engineering strategies, like putting copyright notices at the top of files. Our dataset and evaluation tools are open source which can be contributed to or built upon by future researchers, and will serve as a community resource for evaluating the performance of new models as they emerge. We also describe some good practices for employing LLMs in chemistry. The general success of these models demonstrates that their impact on chemistry teaching and research is poised to be enormous.
more »
« less
- PAR ID:
- 10428078
- Date Published:
- Journal Name:
- Digital Discovery
- Volume:
- 2
- Issue:
- 2
- ISSN:
- 2635-098X
- Page Range / eLocation ID:
- 368 to 376
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
A significant amount of research is focused on developing and evaluating large language models for a variety of code synthesis tasks. These include synthesizing code from natural language, synthesizing tests from code, and synthesizing explanations of code. In contrast, the behavior of instructional code editing with LLMs is understudied. These are tasks in which the model is provided a block of code and an instruction to modify the code. The editing instruction may ask for a feature to be added or removed, describe a bug and ask for a fix, or ask for a different kind of solution. We introduce a carefully crafted benchmark of code editing tasks and use it to evaluate several cutting edge LLMs. Our evaluation exposes a significant gap between the capabilities of state-of-the-art open and closed models. For example, even GPT-3.5-Turbo is better than the best open model at code editing tasks. We also introduce a new, carefully curated, permissively licensed training dataset of code editing tasks coupled with natural language instructions. Using this training dataset, we show that we can fine-tune open Code LLMs to significantly improve their code editing capabilities, closing the gap between open and closed models. All code, data, and models are available at https://github.com/nuprl/CanItEdit.more » « less
-
Large language models (LLMs) are widely used in software development. However, the code generated by LLMs often contains vulnerabilities. Several secure code generation methods have been proposed to address this issue, but their current evaluation schemes leave several concerns unaddressed. Specifically, most existing studies evaluate security and functional correctness separately, using different datasets. That is, they assess vulnerabilities using securityrelated code datasets while validating functionality with general code datasets. In addition, prior research primarily relies on a single static analyzer, CodeQL, to detect vulnerabilities in generated code, which limits the scope of security evaluation. In this work, we conduct a comprehensive study to systematically assess the improvements introduced by four state-of-the-art secure code generation techniques. Specifically, we apply both security inspection and functionality validation to the same generated code and evaluate these two aspects together. We also employ three popular static analyzers and two LLMs to identify potential vulnerabilities in the generated code. Our study reveals that existing techniques often compromise the functionality of generated code to enhance security. Their overall performance remains limited when evaluating security and functionality together. In fact, many techniques even degrade the performance of the base LLM by more than 50%. Our further inspection reveals that these techniques often either remove vulnerable lines of code entirely or generate “garbage code” that is unrelated to the intended task. Moreover, the commonly used static analyzer CodeQL fails to detect several vulnerabilities, further obscuring the actual security improvements achieved by existing techniques. Our study serves as a guideline for a more rigorous and comprehensive evaluation of secure code generation performance in future work.more » « less
-
Understanding and reasoning about code semantics is essential for enhancing code LLMs' abilities to solve real-world software engineering (SE) tasks. Although several code reasoning benchmarks exist, most rely on synthetic datasets or educational coding problems and focus on coarse-grained reasoning tasks such as input/output prediction, limiting their effectiveness in evaluating LLMs in practical SE contexts. To bridge this gap, we propose CodeSense, the first benchmark that makes available a spectrum of fine-grained code reasoning tasks concerned with the software engineering of real-world code. We collected Python, C and Java software projects from real-world repositories. We executed tests from these repositories, collected their execution traces, and constructed a ground truth dataset for fine-grained semantic reasoning tasks. We then performed comprehensive evaluations on state-of-the-art LLMs. Our results show a clear performance gap for the models to handle fine-grained reasoning tasks. Although prompting techniques such as chain-of-thought and in-context learning helped, the lack of code semantics in LLMs fundamentally limit models' capabilities of code reasoning. Besides dataset, benchmark and evaluation, our work produced an execution tracing framework and tool set that make it easy to collect ground truth for fine-grained SE reasoning tasks, offering a strong basis for future benchmark construction and model post training. Our code and data are located at \url{https://codesense-bench.github.io/}.more » « less
-
When seeking information from unfamiliar documents, users frequently pose questions that cannot be answered by the documents. While existing large language models (LLMs) identify these unanswerable questions, they do not assist users in reformulating their questions, thereby reducing their overall utility. We curate CouldAsk, an evaluation benchmark composed of existing and new datasets for document-grounded question answering, specifically designed to study reformulating unanswerable questions. We evaluate state-of-the-art open-source and proprietary LLMs on CouldAsk. The results demonstrate the limited capabilities of these models in reformulating questions. Specifically, GPT-4 and Llama2-7B successfully reformulate questions only 26% and 12% of the time, respectively. Error analysis shows that 62% of the unsuccessful reformulations stem from the models merely rephrasing the questions or even generating identical questions. We publicly release the benchmark and the code to reproduce the experiments.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript
- Journal Article:
- https://doi.org/10.1039/D2DD00087C
