We propose an adaptive fine-tuning algorithm for multimodal large models. The core steps of this algorithm involve two stages of truncation. First, the vast amount of data is projected into a semantic vector space, and the MiniBatchKMeans algorithm is used for automated clustering. This classification ensures that the data within each cluster exhibit high semantic similarity. Next, we process the data in each cluster, calculating the translational difference between the original and perturbed data in the multimodal large model's vector space. This difference serves as a generalization metric for the data. Based on this metric, we select the data with high generalization potential for training. We applied this algorithm to train the InternLM-XComposer2-VL-7B model on two 3090 GPUs using one-third of the GeoChat multimodal remote sensing dataset. The results demonstrate that our algorithm outperforms the state-of-the-art baselines. various baselines. The model trained on our optimally chosen one-third dataset, based on experimental validation, exhibited only 1% reduction in performance across various remote sensing metrics compared to the model trained on the full dataset. This approach significantly preserved general-purpose capabilities while reducing training time by 68.2%. Furthermore, the model achieved scores of 89.86 and 77.19 on the UCMerced and AID evaluation datasets, respectively, surpassing the GeoChat dataset by 5.43 and 5.16 points. It only showed a 0.91-point average decrease on the LRBEN evaluation dataset.

Large language models (LLMs) are renowned for their exceptional capabilities, and applying to a wide range of applications. However, this widespread use brings significant vulnerabilities. Also, it is well observed that there are huge gap which lies in the need for effective detection and mitigation strategies against malicious prompt injection attacks in large language models, as current approaches may not adequately address the complexity and evolving nature of these vulnerabilities in real-world applications. Therefore, this work focuses the impact of malicious prompt injection attacks which is one of most dangerous vulnerability on real LLMs applications. It examines to apply various BERT (Bidirectional Encoder Representations from Transformers) like multilingual BERT, DistilBert for classifying malicious prompts from legitimate prompts. Also, we observed how tokenizing the prompt texts and generating embeddings using multilingual BERT contributes to improve the performance of various machine learning methods: Gaussian Naive Bayes, Random Forest, Support Vector Machine, and Logistic Regression. The performance of each model is rigorously analyzed with various parameters to improve the binary classification to discover malicious prompts. Multilingual BERT approach to embed the prompts significantly improved and outperformed the existing works and achieves an outstanding accuracy of 96.55% by Logistic regression. Additionally, we investigated the incorrect predictions of the model to gain insights into its limitations. The findings can guide researchers in tuning various BERT for finding the most suitable model for diverse LLMs vulnerabilities.

Environmental hazards place certain individuals at disproportionately higher risks. As these hazards increasingly endanger human health, precise identification of the most vulnerable population subgroups is critical for public health. Moderated multiple regression (MMR) offers a straightforward method for investigating this by adding interaction terms between the exposure to a hazard and other population characteristics to a linear regression model. However, when the vulnerabilities are hidden within a cross-section of many characteristics, MMR is often limited in its capabilities to find any meaningful discoveries. Here, we introduce a hybrid method, named regression-guided neural networks (ReGNN), which utilizes artificial neural networks (ANNs) to non-linearly combine predictors, generating a latent representation that interacts with a focal predictor (i.e. variable measuring exposure to an environmental hazard). We showcase the use of ReGNN for investigating the population heterogeneity in the health effects of exposure to air pollution (PM2.5) on cognitive functioning scores. We demonstrate that population heterogeneity that would otherwise be hidden using traditional MMR can be found using ReGNN by comparing its results to the fit results of the traditional MMR models. In essence, ReGNN is a novel tool that enhances traditional regression models by effectively summarizing and quantifying an individual's susceptibility to health risks.

The mean squared error and regularized versions of it are standard loss functions in supervised machine learning. However, calculating these losses for large data sets can be computationally demanding. Modifying an approach of J. Dick and M. Feischl [Journal of Complexity 67 (2021)], we present algorithms to reduce extensive data sets to a smaller size using rank-1 lattices. Rank-1 lattices are quasi-Monte Carlo (QMC) point sets that are, if carefully chosen, well-distributed in a multidimensional unit cube. The compression strategy in the preprocessing step assigns every lattice point a pair of weights depending on the original data and responses, representing its relative importance. As a result, the compressed data makes iterative loss calculations in optimization steps much faster. We analyze the errors of our QMC data compression algorithms and the cost of the preprocessing step for functions whose Fourier coefficients decay sufficiently fast so that they lie in certain Wiener algebras or Korobov spaces. In particular, we prove that our approach can lead to arbitrary high convergence rates as long as the functions are sufficiently smooth.

In tabular biomedical data analysis, tuning models to high accuracy is considered a prerequisite for discussing feature importance, as medical practitioners expect the validity of feature importance to correlate with performance. In this work, we challenge the prevailing belief, showing that low-performing models may also be used for feature importance. We propose experiments to observe changes in feature rank as performance degrades sequentially. Using three synthetic datasets and six real biomedical datasets, we compare the rank of features from full datasets to those with reduced sample sizes (data cutting) or fewer features (feature cutting). In synthetic datasets, feature cutting does not change feature rank, while data cutting shows higher discrepancies with lower performance. In real datasets, feature cutting shows similar or smaller changes than data cutting, though some datasets exhibit the opposite. When feature interactions are controlled by removing correlations, feature cutting consistently shows better stability. By analyzing the distribution of feature importance values and theoretically examining the probability that the model cannot distinguish feature importance between features, we reveal that models can still distinguish feature importance despite performance degradation through feature cutting, but not through data cutting. We conclude that the validity of feature importance can be maintained even at low performance levels if the data size is adequate, which is a significant factor contributing to suboptimal performance in tabular medical data analysis. This paper demonstrates the potential for utilizing feature importance analysis alongside statistical analysis to compare features relatively, even when classifier performance is not satisfactory.

In societies valuing freedom of expression, individuals now frequently express and share their opinions, integrating this practice as a natural part of their routines. Unfortunately, this new social and informational landscape has favored an unprecedented amplification of cyber threats such as hate speech and disinformation, posing significant risks to democratic systems Office of Science and Technology of the Congress of Deputies (Office C) (2023). This situation has intensified and drawn substantial attention from the research community, governmental bodies, and the general public, particularly following extensive disinformation campaigns associated with recent events, including the COVID-19 pandemic Kim and Kesari (2021), the Russia-Ukraine war Pierri et al. (2022), and the Israel-Palestine conflict Aljazeera (2024). Consequently, a structured model encapsulating the key actors, dynamics, and resulting societal impacts is proposed to understand and contextualize the environment being worked on. Figure 1 illustrates our threat model with three main components. In blue, the media and audience as actors in the model, providing the information environment with online news and social network posts that people can read, react to, and comment on. In orange, the content is considered potentially harmful due to intrinsic hateful narratives of today's ecosystem (particularly, public reactions that will be the focus of this research work). In red, the online situation leads to polarization, extremism, and heightened tension, creating a vulnerable environment for society OSMUNDSEN et al. (2021); Cinelli et al. (2021); Pastor-Galindo et al. (2021). In fact, this agitated context serves as a vector for disinformation to become more effective Kim and Kesari (2021).

We introduce jina-embeddings-v3, a novel text embedding model with 570 million parameters, achieves state-of-the-art performance on multilingual data and long-context retrieval tasks, supporting context lengths of up to 8192 tokens. The model includes a set of task-specific Low-Rank Adaptation (LoRA) adapters to generate high-quality embeddings for query-document retrieval, clustering, classification, and text matching. Evaluation on the MTEB benchmark shows that jina-embeddings-v3 outperforms the latest proprietary embeddings from OpenAI and Cohere on English tasks, while achieving superior performance compared to multilingual-e5-large-instruct across all multilingual tasks. With a default output dimension of 1024, users can flexibly reduce the embedding dimensions to as low as 32 without compromising performance, enabled by Matryoshka Representation Learning.

Fairness metrics are used to assess discrimination and disparity of the chances between yellow and blue candidates of getting bias in decision-making processes across various domains, including accepted. Intuitively, we are more certain about the decisions machine learning models and human decision-makers in real-world being made by company A than company B. In the case of company applications. This involves calculating the disparities between probabilistic B, the rejection of blue candidates can be attributed to random outcomes among social groups, such as acceptance rates circumstances. In this case, we would judge company A as more discriminatory between male and female applicants. However, traditional fairness than company B because we are more certain that A metrics do not account for the uncertainty in these processes and is unfair and very uncertain about the unfairness of B. But if both lack of comparability when two decision-makers exhibit the same companies accepted all applicants, the disparity would be 0%, and disparity. Using Bayesian statistics, we quantify the uncertainty of we would conversely judge B as more discriminatory than A. This is the disparity to enhance discrimination assessments. We represent because we are certain that A is fair, while we are uncertain about the each decision-maker, whether a machine learning model or a human, fairness of B. Lastly, when comparing between uncertain fair and uncertain by its disparity and the corresponding uncertainty in that disparity.

Sentiment analysis possesses the potential of diverse applicability on digital platforms. Sentiment analysis extracts the polarity to understand the intensity and subjectivity in the text. This work uses a lexicon-based method to perform sentiment analysis and shows an evaluation of classification models trained over textual data. The lexicon-based methods identify the intensity of emotion and subjectivity at word levels. The categorization identifies the informative words inside a text and specifies the quantitative ranking of the polarity of words. This work is based on a multi-class problem of text being labeled as positive, negative, or neutral. Twitter sentiment dataset containing 1.6 million unprocessed tweets is used with lexicon-based methods like Text Blob and Vader Sentiment to introduce the neutrality measure on text. The analysis of lexicons shows how the word count and the intensity classify the text. A comparative analysis of machine learning models, Naiive Bayes, Support Vector Machines, Multinomial Logistic Regression, Random Forest, and Extreme Gradient (XG) Boost performed across multiple performance metrics. The best estimations are achieved through Random Forest with an accuracy score of 81%. Additionally, sentiment analysis is applied for a personality judgment case against a Twitter profile based on online activity.

In this work, we significantly enhance masked particle modeling (MPM), a self-supervised learning scheme for constructing highly expressive representations of unordered sets relevant to developing foundation models for high-energy physics. In MPM, a model is trained to recover the missing elements of a set, a learning objective that requires no labels and can be applied directly to experimental data. We achieve significant performance improvements over previous work on MPM by addressing inefficiencies in the implementation and incorporating a more powerful decoder. We compare several pre-training tasks and introduce new reconstruction methods that utilize conditional generative models without data tokenization or discretization. We show that these new methods outperform the tokenized learning objective from the original MPM on a new test bed for foundation models for jets, which includes using a wide variety of downstream tasks relevant to jet physics, such as classification, secondary vertex finding, and track identification.