Next-day S&P 500 direction forecasting matters for allocation, hedging, and risk management because broad-index movements transmit quickly across portfolios. Does structured news sentiment help predict next-day S&P 500 direction? We test four feature sets over 2008–2023 in an ablation sequence: technical indicators only (Set A), with FinBERT headline sentiment (Set B), with BERTopic topic-linked sentiment (Set C), and with realized-volatility weighting (Set D). This design makes two contributions: it separates the incremental value of increasingly structured sentiment features, and it tests whether sentiment value is state-dependent across volatility regimes. CatBoost, XGBoost, LightGBM, LSTM, and GRU are evaluated under walk-forward cross-validation, nested cross-validation, and formal statistical tests. On the full sample, sentiment does not deliver a measurable forecasting edge. Walk-forward AUCs sit near 0.50 for every feature set, and pairwise tests find no significant differences. However, this average masks a consistent pattern. Sentiment becomes more informative during high-volatility periods, suggesting that its value is state-dependent rather than uniform. Rolling AUC swings from 0.28 to 0.71 depending on the market period. When we split by VIX regime, Set D reaches 0.5684 AUC during high-volatility episodes ($n=50$, permutation $p=0.213$) while adding almost nothing in calm markets. Set D also has the lowest fold-to-fold variance and the shallowest drawdown in trading simulations. These results imply that the relevant question is not whether sentiment works in general, but when it does. Sentiment does not help on average; whether it helps during stress is suggestive but unconfirmed and needs more crisis-period data to settle.

Any exploit taking advantage of zero-day is called a zero-day attack. Previous research and social media trends show a massive demand for research in zero-day attack detection. This paper analyzes Machine Learning (ML) and Deep Learning (DL) based approaches to create Intrusion Detection Systems (IDS) and scrutinizing them using Explainable AI (XAI) by training an explainer based on randomly sampled data from the testing set. The focus is on using the KDD99 dataset, which has the most research done among all the datasets for detecting zero-day attacks. The paper aims to synthesize the dataset to have fewer classes for multi-class classification, test ML and DL approaches on pattern recognition, establish the robustness and dependability of the model, and establish the interpretability and scalability of the model. We evaluated the performance of four multilayer perceptron (MLP) trained on the KDD99 dataset, including baseline ML models, weighted ML models, truncated ML models, and weighted truncated ML models. Our results demonstrate that the truncated ML model achieves the highest accuracy (99.62%), precision, and recall, while weighted truncated ML model shows lower accuracy (97.26%) but better class representation (less bias) among all the classes with improved unweighted recall score. We also used Shapely Additive exPlanations (SHAP) to train explainer for our truncated models to check for feature importance among the two weighted and unweighted models.

Vision–Language Navigation (VLN) requires embodied agents to interpret natural-language instructions in 3D environments. While significant progress has been made, state-of-the-art agents often struggle with instructions that demand fine-grained visual understanding, correctly grounding specific object attributes, distinguishing between similar objects, and interpreting complex spatial relationships. Many methods rely on static visual features or on global cross-modal attention and lack mechanisms to dynamically resolve ambiguities during navigation. We propose DyGRAV (Dynamic Grounding and Relational Reasoning for VLN), which is a dynamic, uncertainty-aware framework that augments a strong VLN backbone with two adaptively triggered reasoning modules. The Contextual VLM Querying module detects grounding ambiguity for descriptive phrases (e.g., “the ceramic bowl”) and issues targeted queries to a vision–language model (e.g., CLIP) using current visual evidence, while the Incremental Relational Scene Graph (IRSG) constructs lightweight, task-relevant relational memory online to resolve complex spatial relations (e.g., “the chair next to the window”). A budget-aware gating mechanism regulates when these modules are invoked, optimizing the balance between reasoning quality and computational cost through value-per-cost selection and hard budget constraints. Comprehensive evaluations on R2R and RxR, along with a fine-grained diagnostic benchmark emphasizing attribute and relation understanding, show that DyGRAV achieves superior navigation accuracy on challenging instructions while maintaining bounded per-step compute budgets with 97.33% episode-level adherence.