查看更多>>摘要：The powerful ability of ChatGPT has caused widespread concern in the academic community. Malicious users could synthesize dummy academic content through ChatGPT, which is extremely harmful to academic rigor and originality. The need to develop ChatGPT-written content detection algorithms calls for large-scale datasets. In this paper, we initially investigate the possible negative impact of ChatGPT on academia, and present a large-scale CHatGPT-writtEn AbsTract dataset (CHEAT) to support the development of detection algorithms. In particular, the ChatGPT-written abstract dataset contains 35,304 synthetic abstracts, with $Generation$, $Polish$, and $Fusion$ as prominent representatives. Based on these data, we perform a thorough analysis of the existing text synthesis detection algorithms. We show that ChatGPT-written abstracts are detectable with well-trained detectors, while the detection difficulty increases with more human guidance involved.
查看更多>>摘要：Text data augmentation is an effective strategy for overcoming the challenge of limited sample sizes in many natural language processing (NLP) tasks. This challenge is especially prominent in the few-shot learning (FSL) scenario, where the data in the target domain is generally much scarcer and of lowered quality. A natural and widely used strategy to mitigate such challenges is to perform data augmentation to better capture data invariance and increase the sample size. However, current text data augmentation methods either can’t ensure the correct labeling of the generated data (lacking faithfulness), or can’t ensure sufficient diversity in the generated data (lacking compactness), or both. Inspired by the recent success of large language models (LLM), especially the development of ChatGPT, we propose a text data augmentation approach based on ChatGPT (named ”AugGPT”). AugGPT rephrases each sentence in the training samples into multiple conceptually similar but semantically different samples. The augmented samples can then be used in downstream model training. Experiment results on multiple few-shot learning text classification tasks show the superior performance of the proposed AugGPT approach over state-of-the-art text data augmentation methods in terms of testing accuracy and distribution of the augmented samples.
查看更多>>摘要：Whether viewed as an expert or as a source of ‘knowledge hallucination’, the use of ChatGPT in medical practice has stirred ongoing debate. This study sought to evaluate ChatGPT's capabilities in the field of clinical genetics, focusing on tasks such as ‘Clinical genetics exams’, ‘Associations between genetic diseases and pathogenic genes’, and ‘Limitations and trends in clinical genetics’. Results indicated that ChatGPT performed exceptionally well in question-answering tasks, particularly in clinical genetics exams and diagnosing single-gene diseases. It also effectively outlined the current limitations and prospective trends in clinical genetics. However, ChatGPT struggled to provide comprehensive answers regarding multi-gene or epigenetic diseases, particularly with respect to genetic variations or chromosomal abnormalities. In terms of systematic summarization and inference, some randomness was evident in ChatGPT's responses. In summary, while ChatGPT possesses a foundational understanding of general knowledge in clinical genetics due to hyperparameter learning, it encounters significant challenges when delving into specialized knowledge and navigating the complexities of clinical genetics, particularly in mitigating ‘Knowledge Hallucination’. To optimize its performance and depth of expertise in clinical genetics, integration with specialized knowledge databases and knowledge graphs is imperative.
查看更多>>摘要：Large Vision-Language Models (LVLMs) have made significant strides in various multimodal tasks. Notably, GPT4V, Claude, Gemini, and others showcase exceptional multimodal capabilities, marked by profound comprehension and reasoning skills. This study introduces a comprehensive and efficient evaluation framework, TinyLVLM-eHub, to assess LVLMs’ performance, including proprietary models. TinyLVLM-eHub covers six key multimodal capabilities, such as visual perception, knowledge acquisition, reasoning, commonsense understanding, object hallucination, and embodied intelligence. The benchmark, utilizing 2.1K image-text pairs, provides a user-friendly and accessible platform for LVLM evaluation. The evaluation employs the ChatGPT Ensemble Evaluation (CEE) method, which improves alignment with human evaluation compared to word-matching approaches. Results reveal that closed-source API models like GPT4V and GeminiPro-V excel in most capabilities compared to previous open-source LVLMs, though they show some vulnerability in object hallucination. This evaluation underscores areas for LVLM improvement in real-world applications and serves as a foundational assessment for future multimodal advancements.
查看更多>>摘要：Heart failure (HF) poses a significant public health challenge, with a rising global mortality rate. Early detection and prevention of HF could significantly reduce its impact. We introduce a novel methodology for predicting HF risk using 12-lead electrocardiograms (ECGs). We present a novel, lightweight dual attention ECG network designed to capture complex ECG features essential for early HF risk prediction, despite the notable imbalance between low and high-risk groups. This network incorporates a cross-lead attention module and 12 lead-specific temporal attention modules, focusing on cross-lead interactions and each lead's local dynamics. To further alleviate model overfitting, we leverage a large language model (LLM) with a public ECG-Report dataset for pretraining on an ECG-Report alignment task. The network is then fine-tuned for HF risk prediction using two specific cohorts from the U.K. Biobank study, focusing on patients with hypertension (UKB-HYP) and those who have had a myocardial infarction (UKB-MI). The results reveal that LLM-informed pre-training substantially enhances HF risk prediction in these cohorts. The dual attention design not only improves interpretability but also predictive accuracy, outperforming existing competitive methods with C-index scores of 0.6349 for UKB-HYP and 0.5805 for UKB-MI. This demonstrates our method's potential in advancing HF risk assessment with clinical complex ECG data.
查看更多>>摘要：Most existing methods for predicting drug-drug interactions (DDI) predominantly concentrate on capturing the explicit relationships among drugs, overlooking the valuable implicit correlations present between drug pairs (DPs), which leads to weak predictions. To address this issue, this paper introduces a hierarchical multi-relational graph representation learning (HMGRL) approach. Within the framework of HMGRL, we leverage a wealth of drug-related heterogeneous data sources to construct heterogeneous graphs, where nodes represent drugs and edges denote clear and various associations. The relational graph convolutional network (RGCN) is employed to capture diverse explicit relationships between drugs from these heterogeneous graphs. Additionally, a multi-view differentiable spectral clustering (MVDSC) module is developed to capture multiple valuable implicit correlations between DPs. Within the MVDSC, we utilize multiple DP features to construct graphs, where nodes represent DPs and edges denote different implicit correlations. Subsequently, multiple DP representations are generated through graph cutting, each emphasizing distinct implicit correlations. The graph-cutting strategy enables our HMGRL to identify strongly connected communities of graphs, thereby reducing the fusion of irrelevant features. By combining every representation view of a DP, we create high-level DP representations for predicting DDIs. Two genuine datasets spanning three distinct tasks are adopted to gauge the efficacy of our HMGRL. Experimental outcomes unequivocally indicate that HMGRL surpasses several leading-edge methods in performance.
Luis Ibañez-LissenLorena González-ManzanoJosé M. de FuentesManuel Goyanes...
976-987页
查看更多>>摘要：Fake content is a noteworthy threat which is managed by assorted means. This is a serious problem for online shopping platforms whose products can be affected by negative or positive reviews. Artificial intelligence is commonly applied for fake review generation, being transfer learning a promising approach to reduce training requirements. However, the feasibility of generating in-context fake reviews using transfer learning has not been explored yet. This paper analyses the suitability of a couple of transformers (T5 and BART) to generate realistic in-context fake reviews. Results show that 1) the diversity of generated reviews is comparable to existing works; 2) human-based detection is close to random; 3) just reviews generated with one of the used transformers can be detected with 38% precision; and 1 h of training and 8 k real reviews are needed to produce realistic fake reviews.
查看更多>>摘要：Natural terrain scene images play important roles in the geographical research and application. However, it is challenging to collect a large set of terrain scene images. Recently, great progress has been made in image generation. Although impressive results can be achieved, the efficiency of the state-of-the-art methods, e.g., the Vector Quantized Generative Adversarial Network (VQGAN), is still dissatisfying. The VQGAN confronts two issues, i.e., high space complexity and heavy computational demand. To efficiently fulfill the terrain scene generation task, we first collect a Natural Terrain Scene Data Set (NTSD), which contains 36,672 images divided into 38 classes. Then we propose a Lightweight VQGAN (Lit-VQGAN), which uses the fewer parameters and has the lower computational complexity, compared with the VQGAN. A lightweight super-resolution network is further adopted, to speedily derive a high-resolution image from the image that the Lit-VQGAN generates. The Lit-VQGAN can be trained and tested on the NTSD. To our knowledge, either the NTSD or the Lit-VQGAN has not been exploited before.1 Experimental results show that the Lit-VQGAN is more efficient and effective than the VQGAN for the image generation task. These promising results should be due to the lightweight yet effective networks that we design.
Mohammad NadeemShahab Saquib SohailDag Øivind MadsenAhmed Ibrahim Alzahrani...
1001-1012页
查看更多>>摘要：Recent years have witnessed tremendous advancements in Al tools (e.g., ChatGPT, GPT-4, and Bard), driven by the growing power, reasoning, and efficiency of Large Language Models (LLMs). LLMs have been shown to excel in tasks ranging from poem writing and coding to essay generation and puzzle solving. Despite their proficiency in general queries, specialized tasks such as metaphor understanding and fake news detection often require finely tuned models, posing a comparison challenge with specialized Deep Learning (DL). We propose an assessment framework to compare task-specific intelligence with general-purpose LLMs on suicide and depression tendency identification. For this purpose, we trained two DL models on a suicide and depression detection dataset, followed by testing their performance on a test set. Afterward, the same test dataset is used to evaluate the performance of four LLMs (GPT-3.5, GPT-4, Google Bard, and MS Bing) using four classification metrics. The BERT-based DL model performed the best among all, with a testing accuracy of 94.61%, while GPT-4 was the runner-up with accuracy 92.5%. Results demonstrate that LLMs do not outperform the specialized DL models but are able to achieve comparable performance, making them a decent option for downstream tasks without specialized training. However, LLMs outperformed specialized models on the reduced dataset.
NETL
NSTL
IEEE