The observations documented in Cyber Threat Intelligence (CTI) reports play a critical role in describing adversarial behaviors, providing valuable insights for security practitioners to respond to evolving threats. Recent advancements of Large Language Models (LLMs) have demonstrated significant potential in various cybersecurity applications, including CTI report understanding and attack knowledge graph construction. While previous works have proposed benchmarks that focus on the CTI extraction ability of LLMs, the sequential characteristic of adversarial behaviors within CTI reports remains largely unexplored, which holds considerable significance in developing a comprehensive understanding of how adversaries operate. To address this gap, we introduce AttackSeqBench, a benchmark tailored to systematically evaluate LLMs' capability to understand and reason attack sequences in CTI reports. Our benchmark encompasses three distinct Question Answering (QA) tasks, each task focuses on the varying granularity in adversarial behavior. To alleviate the laborious effort of QA construction, we carefully design an automated dataset construction pipeline to create scalable and well-formulated QA datasets based on real-world CTI reports. To ensure the quality of our dataset, we adopt a hybrid approach of combining human evaluation and systematic evaluation metrics. We conduct extensive experiments and analysis with both fast-thinking and slow-thinking LLMs, while highlighting their strengths and limitations in analyzing the sequential patterns in cyber attacks. The overarching goal of this work is to provide a benchmark that advances LLM-driven CTI report understanding and fosters its application in real-world cybersecurity operations. Our dataset and code are available at https://github.com/Javiery3889/AttackSeqBench .

The reliability of large language models remains a critical challenge, particularly due to their susceptibility to hallucinations and factual inaccuracies during text generation. Existing solutions either underutilize models' self-correction with preemptive strategies or use costly post-hoc verification. To further explore the potential of real-time self-verification and correction, we present Dynamic Self-Verify Decoding (DSVD), a novel decoding framework that enhances generation reliability through real-time hallucination detection and efficient error correction. DSVD integrates two key components: (1) parallel self-verification architecture for continuous quality assessment, (2) dynamic rollback mechanism for targeted error recovery. Extensive experiments across five benchmarks demonstrate DSVD's effectiveness, achieving significant improvement in truthfulness (Quesetion-Answering) and factual accuracy (FActScore). Results show the DSVD can be further incorporated with existing faithful decoding methods to achieve stronger performance. Our work establishes that real-time self-verification during generation offers a viable path toward more trustworthy language models without sacrificing practical deployability.

Modeling environmental ecosystems is essential for effective resource management, sustainable development, and understanding complex ecological processes. However, traditional methods frequently struggle with the inherent complexity, interconnectedness, and limited data of such systems. Foundation models, with their large-scale pre-training and universal representations, offer transformative opportunities by integrating diverse data sources, capturing spatiotemporal dependencies, and adapting to a broad range of tasks. This survey presents a comprehensive overview of foundation model applications in environmental science, highlighting advancements in forward prediction, data generation, data assimilation, downscaling, model ensembling, and decision-making across domains. We also detail the development process of these models, covering data collection, architecture design, training, tuning, and evaluation. By showcasing these emerging methods, we aim to foster interdisciplinary collaboration and advance the integration of cutting-edge machine learning for sustainable solutions in environmental science.

Recently, provenance-based intrusion detection systems (PIDSes) have been widely proposed for endpoint threat analysis. However, due to the lack of systematic integration and utilization of knowledge, existing PIDSes still require significant manual intervention for practical deployment, making full automation challenging. This paper presents a disruptive innovation by categorizing PIDSes according to the types of knowledge they utilize. In response to the prevalent issue of ``knowledge silos problem'' in existing research, we introduce a novel knowledge-driven provenance-based intrusion detection framework, powered by large language models (LLMs). We also present OmniSec, a best practice system built upon this framework. By integrating attack representation knowledge, threat intelligence knowledge, and benign behavior knowledge, OmniSec outperforms the state-of-the-art approaches on public benchmark datasets. OmniSec is available online at https://anonymous.4open.science/r/PIDS-with-LLM-613B.

Machine learning models have shown great success in predicting weather up to two weeks ahead, outperforming process-based benchmarks. However, existing approaches mostly focus on the prediction task, and do not incorporate the necessary data assimilation. Moreover, these models suffer from error accumulation in long roll-outs, limiting their applicability to seasonal predictions or climate projections. Here, we introduce Generative Assimilation and Prediction (GAP), a unified deep generative framework for assimilation and prediction of both weather and climate. By learning to quantify the probabilistic distribution of atmospheric states under observational, predictive, and external forcing constraints, GAP excels in a broad range of weather-climate related tasks, including data assimilation, seamless prediction, and climate simulation. In particular, GAP is competitive with state-of-the-art ensemble assimilation, probabilistic weather forecast and seasonal prediction, yields stable millennial simulations, and reproduces climate variability from daily to decadal time scales.

With the rapid growth of IoT devices, ensuring robust network security has become a critical challenge. Traditional intrusion detection systems (IDSs) often face limitations in detecting sophisticated attacks within high-dimensional and complex data environments. This paper presents a novel approach to network anomaly detection using hyperdimensional computing (HDC) techniques, specifically applied to the NSL-KDD dataset. The proposed method leverages the efficiency of HDC in processing large-scale data to identify both known and unknown attack patterns. The model achieved an accuracy of 91.55% on the KDDTrain+ subset, outperforming traditional approaches. These comparative evaluations underscore the model's superior performance, highlighting its potential in advancing anomaly detection for IoT networks and contributing to more secure and intelligent cybersecurity solutions.

In this work, we introduce the Equivariance Seeker Model (ESM), a data-driven method for discovering the underlying finite equivariant symmetry group of an arbitrary function. ESM achieves this by optimizing a loss function that balances equivariance preservation with the penalization of redundant solutions, ensuring the complete and accurate identification of all symmetry transformations. We apply this framework specifically to dynamical systems, identifying their symmetry groups directly from observed trajectory data. To demonstrate its versatility, we test ESM on multiple systems in two distinct scenarios: (i) when the governing equations are known theoretically and (ii) when they are unknown, and the equivariance finding relies solely on observed data. The latter case highlights ESM's fully data-driven capability, as it requires no prior knowledge of the system's equations to operate.

Predicting clinical outcomes from preclinical data is essential for identifying safe and effective drug combinations. Current models rely on structural or target-based features to identify high-efficacy, low-toxicity drug combinations. However, these approaches fail to incorporate the multimodal data necessary for accurate, clinically-relevant predictions. Here, we introduce MADRIGAL, a multimodal AI model that learns from structural, pathway, cell viability, and transcriptomic data to predict drug combination effects across 953 clinical outcomes and 21842 compounds, including combinations of approved drugs and novel compounds in development. MADRIGAL uses a transformer bottleneck module to unify preclinical drug data modalities while handling missing data during training and inference--a major challenge in multimodal learning. It outperforms single-modality methods and state-of-the-art models in predicting adverse drug interactions. MADRIGAL performs virtual screening of anticancer drug combinations and supports polypharmacy management for type II diabetes and metabolic dysfunction-associated steatohepatitis (MASH). It identifies transporter-mediated drug interactions. MADRIGAL predicts resmetirom, the first and only FDA-approved drug for MASH, among therapies with the most favorable safety profile. It supports personalized cancer therapy by integrating genomic profiles from cancer patients. Using primary acute myeloid leukemia samples and patient-derived xenograft models, it predicts the efficacy of personalized drug combinations. Integrating MADRIGAL with a large language model allows users to describe clinical outcomes in natural language, improving safety assessment by identifying potential adverse interactions and toxicity risks. MADRIGAL provides a multimodal approach for designing combination therapies with improved predictive accuracy and clinical relevance.

In recent years, Large Language Models have attracted growing interest for their significant potential, though concerns have rapidly emerged regarding unsafe behaviors stemming from inherent stereotypes and biases. Most research on stereotypes in LLMs has primarily relied on indirect evaluation setups, in which models are prompted to select between pairs of sentences associated with particular social groups. Recently, direct evaluation methods have emerged, examining open-ended model responses to overcome limitations of previous approaches, such as annotator biases. Most existing studies have focused on English-centric LLMs, whereas research on non-English models, particularly Japanese, remains sparse, despite the growing development and adoption of these models. This study examines the safety of Japanese LLMs when responding to stereotype-triggering prompts in direct setups. We constructed 3,612 prompts by combining 301 social group terms, categorized by age, gender, and other attributes, with 12 stereotype-inducing templates in Japanese. Responses were analyzed from three foundational models trained respectively on Japanese, English, and Chinese language. Our findings reveal that LLM-jp, a Japanese native model, exhibits the lowest refusal rate and is more likely to generate toxic and negative responses compared to other models. Additionally, prompt format significantly influence the output of all models, and the generated responses include exaggerated reactions toward specific social groups, varying across models. These findings underscore the insufficient ethical safety mechanisms in Japanese LLMs and demonstrate that even high-accuracy models can produce biased outputs when processing Japanese-language prompts. We advocate for improving safety mechanisms and bias mitigation strategies in Japanese LLMs, contributing to ongoing discussions on AI ethics beyond linguistic boundaries.

The increased adoption of Large Language Models (LLMs) and their potential to shape public opinion have sparked interest in assessing these models' political leanings. Building on previous research that compared LLMs and human opinions and observed political bias in system responses, we take a step further to investigate the underlying causes of such biases by empirically examining how the values and biases embedded in training corpora shape model outputs. Specifically, we propose a method to quantitatively evaluate political leanings embedded in the large pretraining corpora. Subsequently we investigate to whom are the LLMs' political leanings more aligned with, their pretrainig corpora or the surveyed human opinions. As a case study, we focus on probing the political leanings of LLMs in 32 U.S. Supreme Court cases, addressing contentious topics such as abortion and voting rights. Our findings reveal that LLMs strongly reflect the political leanings in their training data, and no strong correlation is observed with their alignment to human opinions as expressed in surveys. These results underscore the importance of responsible curation of training data and the need for robust evaluation metrics to ensure LLMs' alignment with human-centered values.