**《Science》期刊引入Proofig AI技术，强化学术图像真实性检测**
全球知名科研期刊《Science》近日宣布，为打击学术造假，已正式引入了一款名为Proofig的先进人工智能软件，以自动检测并预防论文中可能出现的图像欺诈行为。据《Science》期刊编辑团队透露，他们在过去数月内对Proofig进行了详尽的测试，结果令人满意。
Proofig AI软件的突出功能在于其能够智能识别并标记出经过篡改的图像，包括那些被裁剪、旋转或调整过对比度的图片。这一技术的应用将极大地提升期刊在审稿过程中对图像真实性的把控能力，帮助编辑团队快速定位可能存在欺诈问题的图像，以便进行进一步的人工审核。
学术造假是科研界长期以来面临的严峻问题，它不仅损害了科研的公正性，也对全球的知识积累和科技进步造成了负面影响。《Science》期刊此次引入AI技术，旨在通过技术手段提升学术诚信，保障科研成果的可靠性。此举表明，科技期刊正在积极拥抱新技术，以应对日益复杂的学术不端行为。
《Science》的这一创新举措为全球科研期刊树立了新的标杆，预示着未来学术出版领域在防止学术欺诈方面将有更高效、更智能的解决方案。通过AI辅助审核，有望在更大程度上维护学术界的公正与透明，保护科研的纯净土壤。
英语如下：
**News Title:** “Science Journal Employs Proofig AI to Combat Academic Fraud and Uphold Research Integrity”
**Keywords:** Science Journal, Proofig AI, Image Fraud Detection
**News Content:**
The renowned scientific journal *Science* has recently announced the adoption of Proofig, an advanced artificial intelligence software, to strengthen its detection of manipulated academic images in a bid to combat fraud. The journal’s editorial team has disclosed that they have extensively tested Proofig over the past few months with satisfactory results.
The standout feature of Proofig AI lies in its ability to intelligently identify and flag altered images, including those that have been cropped, rotated, or had their contrast adjusted. This technology will significantly enhance the journal’s capacity to assess image authenticity during the review process, enabling the editorial team to swiftly pinpoint potential fraudulent images for further manual scrutiny.
Academic fraud has long been a significant challenge in the research community, undermining the fairness of research and negatively impacting global knowledge accumulation and scientific progress. By incorporating AI technology, *Science* aims to boost academic integrity and ensure the reliability of research outcomes. This move signals that scientific journals are embracing new technologies to tackle the increasingly sophisticated instances of academic misconduct.
*Science*’s innovative step sets a new standard for global research journals and foreshadows more efficient and intelligent solutions in the future to prevent academic fraud within the scholarly publishing landscape. With AI-assisted review, there is a promising prospect for upholding fairness and transparency in academia, safeguarding the purity of the research domain.
【来源】https://www.ithome.com/0/743/509.htm
Views: 1