Regulatory compliance question answering (QA) requires precise, verifiable information, and domain-specific expertise, posing challenges for Large Language Models (LLMs). In this work, we present a novel multi-agent framework that integrates a Knowledge Graph (KG) of Regulatory triplets with Retrieval-Augmented Generation (RAG) to address these demands. First, agents build and maintain an ontology-free KG by extracting subject--predicate--object (SPO) triplets from regulatory documents and systematically cleaning, normalizing, deduplicating, and updating them. Second, these triplets are embedded and stored along with their corresponding textual sections and metadata in a single enriched vector database, allowing for both graph-based reasoning and efficient information retrieval. Third, an orchestrated agent pipeline leverages triplet-level retrieval for question answering, ensuring high semantic alignment between user queries and the factual "who-did-what-to-whom" core captured by the graph. Our hybrid system outperforms conventional methods in complex regulatory queries, ensuring factual correctness with embedded triplets, enabling traceability through a unified vector database, and enhancing understanding through subgraph visualization, providing a robust foundation for compliance-driven and broader audit-focused applications.

The field of explainable natural language processing (NLP) has grown rapidly in recent years. The growing opacity of complex models calls for transparency and explanations of their decisions, which is crucial to understand their reasoning and facilitate deployment, especially in high-stakes environments. Despite increasing attention given to explainable NLP, practitioners' perspectives regarding its practical adoption and effectiveness remain underexplored. This paper addresses this research gap by investigating practitioners' experiences with explainability methods, specifically focusing on their motivations for adopting such methods, the techniques employed, satisfaction levels, and the practical challenges encountered in real-world NLP applications. Through a qualitative interview-based study with industry practitioners and complementary interviews with academic researchers, we systematically analyze and compare their perspectives. Our findings reveal conceptual gaps, low satisfaction with current explainability methods, and highlight evaluation challenges. Our findings emphasize the need for clear definitions and user-centric frameworks for better adoption of explainable NLP in practice.

As Large Language Models (LLMs) become increasingly autonomous and integrated into critical societal functions, the focus of AI safety must evolve from mitigating harmful content to evaluating underlying behavioral alignment. Current safety benchmarks do not systematically probe a model's decision-making in scenarios where its own instrumental goals - such as self-preservation, resource acquisition, or goal completion - conflict with human safety. This represents a critical gap in our ability to measure and mitigate risks associated with emergent, misaligned behaviors. To address this, we introduce PacifAIst (Procedural Assessment of Complex Interactions for Foundational Artificial Intelligence Scenario Testing), a focused benchmark of 700 challenging scenarios designed to quantify self-preferential behavior in LLMs. The benchmark is structured around a novel taxonomy of Existential Prioritization (EP), with subcategories testing Self-Preservation vs. Human Safety (EP1), Resource Conflict (EP2), and Goal Preservation vs. Evasion (EP3). We evaluated eight leading LLMs. The results reveal a significant performance hierarchy. Google's Gemini 2.5 Flash achieved the highest Pacifism Score (P-Score) at 90.31%, demonstrating strong human-centric alignment. In a surprising result, the much-anticipated GPT-5 recorded the lowest P-Score (79.49%), indicating potential alignment challenges. Performance varied significantly across subcategories, with models like Claude Sonnet 4 and Mistral Medium struggling notably in direct self-preservation dilemmas. These findings underscore the urgent need for standardized tools like PacifAIst to measure and mitigate risks from instrumental goal conflicts, ensuring future AI systems are not only helpful in conversation but also provably "pacifist" in their behavioral priorities.

There have been numerous studies evaluating bias of LLMs towards political topics. However, how positions towards these topics in model outputs are highly sensitive to the prompt. What happens when the prompt itself is suggestive of certain arguments towards those positions remains underexplored. This is crucial for understanding how robust these bias evaluations are and for understanding model behaviour, as these models frequently interact with opinionated text. To that end, we conduct experiments for political bias evaluation in presence of supporting and refuting arguments. Our experiments show that such arguments substantially alter model responses towards the direction of the provided argument in both single-turn and multi-turn settings. Moreover, we find that the strength of these arguments influences the directional agreement rate of model responses. These effects point to a sycophantic tendency in LLMs adapting their stance to align with the presented arguments which has downstream implications for measuring political bias and developing effective mitigation strategies.

Explainable Artificial Intelligence (XAI) techniques, such as SHapley Additive exPlanations (SHAP), have become essential tools for interpreting complex ensemble tree-based models, especially in high-stakes domains such as healthcare analytics. However, SHAP values are usually treated as point estimates, which disregards the inherent and ubiquitous uncertainty in predictive models and data. This uncertainty has two primary sources: aleatoric and epistemic. The aleatoric uncertainty, which reflects the irreducible noise in the data. The epistemic uncertainty, which arises from a lack of data. In this work, we propose an approach for decomposing uncertainty in SHAP values into aleatoric, epistemic, and entanglement components. This approach integrates Dempster-Shafer evidence theory and hypothesis sampling via Dirichlet processes over tree ensembles. We validate the method across three real-world use cases with descriptive statistical analyses that provide insight into the nature of epistemic uncertainty embedded in SHAP explanations. The experimentations enable to provide more comprehensive understanding of the reliability and interpretability of SHAP-based attributions. This understanding can guide the development of robust decision-making processes and the refinement of models in high-stakes applications. Through our experiments with multiple datasets, we concluded that features with the highest SHAP values are not necessarily the most stable. This epistemic uncertainty can be reduced through better, more representative data and following appropriate or case-desired model development techniques. Tree-based models, especially bagging, facilitate the effective quantification of epistemic uncertainty.

Curb ramps are critical for urban accessibility, but robustly detecting them in images remains an open problem due to the lack of large-scale, high-quality datasets. While prior work has attempted to improve data availability with crowdsourced or manually labeled data, these efforts often fall short in either quality or scale. In this paper, we introduce and evaluate a two-stage pipeline called RampNet to scale curb ramp detection datasets and improve model performance. In Stage 1, we generate a dataset of more than 210,000 annotated Google Street View (GSV) panoramas by auto-translating government-provided curb ramp location data to pixel coordinates in panoramic images. In Stage 2, we train a curb ramp detection model (modified ConvNeXt V2) from the generated dataset, achieving state-of-the-art performance. To evaluate both stages of our pipeline, we compare to manually labeled panoramas. Our generated dataset achieves 94.0% precision and 92.5% recall, and our detection model reaches 0.9236 AP -- far exceeding prior work. Our work contributes the first large-scale, high-quality curb ramp detection dataset, benchmark, and model.

Low-mass stars and brown dwarfs -- spectral types (SpTs) M0 and later -- play a significant role in studying stellar and substellar processes and demographics, reaching down to planetary-mass objects. Currently, the classification of these sources remains heavily reliant on visual inspection of spectral features, equivalent width measurements, or narrow-/wide-band spectral indices. Recent advances in machine learning (ML) methods offer automated approaches for spectral typing, which are becoming increasingly important as large spectroscopic surveys such as Gaia, SDSS, and SPHEREx generate datasets containing millions of spectra. We investigate the application of ML in spectral type classification on low-resolution (R $\sim$ 120) near-infrared spectra of M0--T9 dwarfs obtained with the SpeX instrument on the NASA Infrared Telescope Facility. We specifically aim to classify the gravity- and metallicity-dependent subclasses for late-type dwarfs. We used binned fluxes as input features and compared the efficacy of spectral type estimators built using Random Forest (RF), Support Vector Machine (SVM), and K-Nearest Neighbor (KNN) models. We tested the influence of different normalizations and analyzed the relative importance of different spectral regions for surface gravity and metallicity subclass classification. Our best-performing model (using KNN) classifies 95.5 $\pm$ 0.6% of sources to within $\pm$1 SpT, and assigns surface gravity and metallicity subclasses with 89.5 $\pm$ 0.9% accuracy. We test the dependence of signal-to-noise ratio on classification accuracy and find sources with SNR $\gtrsim$ 60 have $\gtrsim$ 95% accuracy. We also find that zy-band plays the most prominent role in the RF model, with FeH and TiO having the highest feature importance.

This paper examines Modern Mercantilism, characterized by rising economic nationalism, strategic technological decoupling, and geopolitical fragmentation, as a disruptive shift from the post-1945 globalization paradigm. It applies Principal Component Analysis (PCA) to 768-dimensional SBERT-generated semantic embeddings of curated news articles to extract orthogonal latent factors that discriminate binary event outcomes linked to protectionism, technological sovereignty, and bloc realignments. Analysis of principal component loadings identifies key semantic features driving classification performance, enhancing interpretability and predictive accuracy. This methodology provides a scalable, data-driven framework for quantitatively tracking emergent mercantilist dynamics through high-dimensional text analytics

The field of Explainable AI (XAI) offers a wide range of techniques for making complex models interpretable. Yet, in practice, generating meaningful explanations is a context-dependent task that requires intentional design choices to ensure accessibility and transparency. This paper reframes explanation as a situated design process -- an approach particularly relevant for practitioners involved in building and deploying explainable systems. Drawing on prior research and principles from design thinking, we propose a three-part framework for explanation design in XAI: asking Who needs the explanation, What they need explained, and How that explanation should be delivered. We also emphasize the need for ethical considerations, including risks of epistemic inequality, reinforcing social inequities, and obscuring accountability and governance. By treating explanation as a sociotechnical design process, this framework encourages a context-aware approach to XAI that supports effective communication and the development of ethically responsible explanations.

Vision Language Model (VLM)-based agents are stateful, autonomous entities capable of perceiving and interacting with their environments through vision and language. Multi-agent systems comprise specialized agents who collaborate to solve a (complex) task. A core security property is robustness, stating that the system should maintain its integrity under adversarial attacks. However, the design of existing multi-agent systems lacks the robustness consideration, as a successful exploit against one agent can spread and infect other agents to undermine the entire system's assurance. To address this, we propose a new defense approach, Cowpox, to provably enhance the robustness of multi-agent systems. It incorporates a distributed mechanism, which improves the recovery rate of agents by limiting the expected number of infections to other agents. The core idea is to generate and distribute a special cure sample that immunizes an agent against the attack before exposure and helps recover the already infected agents. We demonstrate the effectiveness of Cowpox empirically and provide theoretical robustness guarantees.