arXiv:2510.20099v1 Announce Type: new Abstract: We present AI PB, a production-scale generative agent deployed in real retail finance. Unlike reactive chatbots that answer queries passively, AI PB proactively generates grounded, compliant, and user-specific investment insights. It integrates (i) a component-based orchestration layer that deterministically routes between internal and external LLMs based on data sensitivity, (ii) a hybrid retrieval pipeline using OpenSearch and the finance-domain embedding model, and (iii) a multi-stage recommendation mechanism combining rule heuristics, sequential behavioral modeling, and contextual bandits. Operating fully on-premises under Korean financial regulations, the system employs Docker Swarm and vLLM across 24 X NVIDIA H100 GPUs. Through human QA and system metrics, we demonstrate that grounded generation with explicit routing and layered safety can deliver trustworthy AI insights in high-stakes finance.

arXiv:2510.20332v1 Announce Type: new Abstract: Artificial intelligence (AI) holds great promise for transforming healthcare. However, despite significant advances, the integration of AI solutions into real-world clinical practice remains limited. A major barrier is the quality and fairness of training data, which is often compromised by biased data collection practices. This paper draws on insights from the AI4HealthyAging project, part of Spain's national R&D initiative, where our task was to detect biases during clinical data collection. We identify several types of bias across multiple use cases, including historical, representation, and measurement biases. These biases manifest in variables such as sex, gender, age, habitat, socioeconomic status, equipment, and labeling. We conclude with practical recommendations for improving the fairness and robustness of clinical problem design and data collection. We hope that our findings and experience contribute to guiding future projects in the development of fairer AI systems in healthcare.

arXiv:2510.20467v1 Announce Type: new Abstract: Knowledge graph alignment is the task of matching equivalent entities (that is, instances and classes) and relations across two knowledge graphs. Most existing methods focus on pure entity-level alignment, computing the similarity of entities in some embedding space. They lack interpretable reasoning and need training data to work. In this paper, we propose FLORA, a simple yet effective method that (1) is unsupervised, i.e., does not require training data, (2) provides a holistic alignment for entities and relations iteratively, (3) is based on fuzzy logic and thus delivers interpretable results, (4) provably converges, (5) allows dangling entities, i.e., entities without a counterpart in the other KG, and (6) achieves state-of-the-art results on major benchmarks.

arXiv:2510.20568v1 Announce Type: new Abstract: The worlds people have strong opinions about artificial intelligence (AI), and they want policymakers to listen. Governments are inviting public comment on AI, but as they translate input into policy, much of what citizens say is lost. Policymakers are missing a critical opportunity to build trust in AI and its governance. This paper compares three countries, Australia, Colombia, and the United States, that invited citizens to comment on AI risks and policies. Using a landscape analysis, the authors examined how each government solicited feedback and whether that input shaped governance. Yet in none of the three cases did citizens and policymakers establish a meaningful dialogue. Governments did little to attract diverse voices or publicize calls for comment, leaving most citizens unaware or unprepared to respond. In each nation, fewer than one percent of the population participated. Moreover, officials showed limited responsiveness to the feedback they received, failing to create an effective feedback loop. The study finds a persistent gap between the promise and practice of participatory AI governance. The authors conclude that current approaches are unlikely to build trust or legitimacy in AI because policymakers are not adequately listening or responding to public concerns. They offer eight recommendations: promote AI literacy; monitor public feedback; broaden outreach; hold regular online forums; use innovative engagement methods; include underrepresented groups; respond publicly to input; and make participation easier.

arXiv:2510.20448v1 Announce Type: cross Abstract: Drug combinations offer therapeutic benefits but also carry the risk of adverse drug-drug interactions (DDIs), especially under complex molecular structures. Accurate DDI event prediction requires capturing fine-grained inter-drug relationships, which are critical for modeling metabolic mechanisms such as enzyme-mediated competition. However, existing approaches typically rely on isolated drug representations and fail to explicitly model atom-level cross-molecular interactions, limiting their effectiveness across diverse molecular complexities and DDI type distributions. To address these limitations, we propose MolBridge, a novel atom-level joint graph refinement framework for robust DDI event prediction. MolBridge constructs a joint graph that integrates atomic structures of drug pairs, enabling direct modeling of inter-drug associations. A central challenge in such joint graph settings is the potential loss of information caused by over-smoothing when modeling long-range atomic dependencies. To overcome this, we introduce a structure consistency module that iteratively refines node features while preserving the global structural context. This joint design allows MolBridge to effectively learn both local and global interaction outperforms state-of-the-art baselines, achieving superior performance across long-tail and inductive scenarios. patterns, yielding robust representations across both frequent and rare DDI types. Extensive experiments on two benchmark datasets show that MolBridge consistently. These results demonstrate the advantages of fine-grained graph refinement in improving the accuracy, robustness, and mechanistic interpretability of DDI event prediction.This work contributes to Web Mining and Content Analysis by developing graph-based methods for mining and analyzing drug-drug interaction networks.

arXiv:2510.20692v1 Announce Type: cross Abstract: Cloud computing is ubiquitous, with a growing number of services being hosted on the cloud every day. Typical cloud compute systems allow administrators to write policies implementing access control rules which specify how access to private data is governed. These policies must be manually written, and due to their complexity can often be error prone. Moreover, existing policies often implement complex access control specifications and thus can be difficult to precisely analyze in determining their behavior works exactly as intended. Recently, Large Language Models (LLMs) have shown great success in automated code synthesis and summarization. Given this success, they could potentially be used for automatically generating access control policies or aid in understanding existing policies. In this paper, we explore the effectiveness of LLMs for access control policy synthesis and summarization. Specifically, we first investigate diverse LLMs for access control policy synthesis, finding that: although LLMs can effectively generate syntactically correct policies, they have permissiveness issues, generating policies equivalent to the given specification 45.8% of the time for non-reasoning LLMs, and 93.7% of the time for reasoning LLMs. We then investigate how LLMs can be used to analyze policies by introducing a novel semantic-based request summarization approach which leverages LLMs to generate a precise characterization of the requests allowed by a policy. Our results show that while there are significant hurdles in leveraging LLMs for automated policy generation, LLMs show promising results when combined with symbolic approaches in analyzing existing policies.

arXiv:2510.20721v1 Announce Type: cross Abstract: Large language models (LLMs) have seen rapid adoption for tasks such as drafting emails, summarizing meetings, and answering health questions. In such uses, users may need to share private information (e.g., health records, contact details). To evaluate LLMs' ability to identify and redact such private information, prior work developed benchmarks (e.g., ConfAIde, PrivacyLens) with real-life scenarios. Using these benchmarks, researchers have found that LLMs sometimes fail to keep secrets private when responding to complex tasks (e.g., leaking employee salaries in meeting summaries). However, these evaluations rely on LLMs (proxy LLMs) to gauge compliance with privacy norms, overlooking real users' perceptions. Moreover, prior work primarily focused on the privacy-preservation quality of responses, without investigating nuanced differences in helpfulness. To understand how users perceive the privacy-preservation quality and helpfulness of LLM responses to privacy-sensitive scenarios, we conducted a user study with 94 participants using 90 scenarios from PrivacyLens. We found that, when evaluating identical responses to the same scenario, users showed low agreement with each other on the privacy-preservation quality and helpfulness of the LLM response. Further, we found high agreement among five proxy LLMs, while each individual LLM had low correlation with users' evaluations. These results indicate that the privacy and helpfulness of LLM responses are often specific to individuals, and proxy LLMs are poor estimates of how real users would perceive these responses in privacy-sensitive scenarios. Our results suggest the need to conduct user-centered studies on measuring LLMs' ability to help users while preserving privacy. Additionally, future research could investigate ways to improve the alignment between proxy LLMs and users for better estimation of users' perceived privacy and utility.

arXiv:2510.20727v1 Announce Type: cross Abstract: Objective: Fluoropyrimidines are widely prescribed for colorectal and breast cancers, but are associated with toxicities such as hand-foot syndrome and cardiotoxicity. Since toxicity documentation is often embedded in clinical notes, we aimed to develop and evaluate natural language processing (NLP) methods to extract treatment and toxicity information. Materials and Methods: We constructed a gold-standard dataset of 236 clinical notes from 204,165 adult oncology patients. Domain experts annotated categories related to treatment regimens and toxicities. We developed rule-based, machine learning-based (Random Forest, Support Vector Machine [SVM], Logistic Regression [LR]), deep learning-based (BERT, ClinicalBERT), and large language models (LLM)-based NLP approaches (zero-shot and error-analysis prompting). Models used an 80:20 train-test split. Results: Sufficient data existed to train and evaluate 5 annotated categories. Error-analysis prompting achieved optimal precision, recall, and F1 scores (F1=1.000) for treatment and toxicities extraction, whereas zero-shot prompting reached F1=1.000 for treatment and F1=0.876 for toxicities extraction.LR and SVM ranked second for toxicities (F1=0.937). Deep learning underperformed, with BERT (F1=0.873 treatment; F1= 0.839 toxicities) and ClinicalBERT (F1=0.873 treatment; F1 = 0.886 toxicities). Rule-based methods served as our baseline with F1 scores of 0.857 in treatment and 0.858 in toxicities. Discussion: LMM-based approaches outperformed all others, followed by machine learning methods. Machine and deep learning approaches were limited by small training data and showed limited generalizability, particularly for rare categories. Conclusion: LLM-based NLP most effectively extracted fluoropyrimidine treatment and toxicity information from clinical notes, and has strong potential to support oncology research and pharmacovigilance.

arXiv:2510.20733v1 Announce Type: cross Abstract: Natural language has long enabled human cooperation, but its lossy, ambiguous, and indirect nature limits the potential of collective intelligence. While machines are not subject to these constraints, most LLM-based multi-agent systems still rely solely on natural language, exchanging tokens or their embeddings. To go beyond language, we introduce a new paradigm, thought communication, which enables agents to interact directly mind-to-mind, akin to telepathy. To uncover these latent thoughts in a principled way, we formalize the process as a general latent variable model, where agent states are generated by an unknown function of underlying thoughts. We prove that, in a nonparametric setting without auxiliary information, both shared and private latent thoughts between any pair of agents can be identified. Moreover, the global structure of thought sharing, including which agents share which thoughts and how these relationships are structured, can also be recovered with theoretical guarantees. Guided by the established theory, we develop a framework that extracts latent thoughts from all agents prior to communication and assigns each agent the relevant thoughts, along with their sharing patterns. This paradigm naturally extends beyond LLMs to all modalities, as most observational data arise from hidden generative processes. Experiments on both synthetic and real-world benchmarks validate the theory and demonstrate the collaborative advantages of thought communication. We hope this work illuminates the potential of leveraging the hidden world, as many challenges remain unsolvable through surface-level observation alone, regardless of compute or data scale.

arXiv:2510.20800v1 Announce Type: cross Abstract: Recently, Sharma et al. suggested a method called Layer-SElective-Rank reduction (LASER) which demonstrated that pruning high-order components of carefully chosen LLM's weight matrices can boost downstream accuracy -- without any gradient-based fine-tuning. Yet LASER's exhaustive, per-matrix search (each requiring full-dataset forward passes) makes it impractical for rapid deployment. We demonstrate that this overhead can be removed and find that: (i) Only a small, carefully chosen subset of matrices needs to be inspected -- eliminating the layer-by-layer sweep, (ii) The gradient of each matrix's singular values pinpoints which matrices merit reduction, (iii) Increasing the factorization search space by allowing matrices rows to cluster around multiple subspaces and then decomposing each cluster separately further reduces overfitting on the original training data and further lifts accuracy by up to 24.6 percentage points, and finally, (iv) we discover that evaluating on just 100 samples rather than the full training data -- both for computing the indicative gradients and for measuring the final accuracy -- suffices to further reduce the search time; we explain that as adaptation to downstream tasks is dominated by prompting style, not dataset size. As a result, we show that combining these findings yields a fast and robust adaptation algorithm for downstream tasks. Overall, with a single gradient step on 100 examples and a quick scan of the top candidate layers and factorization techniques, we can adapt LLMs to new datasets -- entirely without fine-tuning.

arXiv:2505.14667v4 Announce Type: replace Abstract: Large Reasoning Models (LRMs) have become powerful tools for complex problem solving, but their structured reasoning pathways can lead to unsafe outputs when exposed to harmful prompts. Existing safety alignment methods reduce harmful outputs but can degrade reasoning depth, leading to significant trade-offs in complex, multi-step tasks, and remain vulnerable to sophisticated jailbreak attacks. To address this, we introduce SAFEPATH, a lightweight alignment method that fine-tunes LRMs to emit a short, 8-token Safety Primer at the start of their reasoning, in response to harmful prompts, while leaving the rest of the reasoning process unsupervised. Empirical results across multiple benchmarks indicate that SAFEPATH effectively reduces harmful outputs while maintaining reasoning performance. Specifically, SAFEPATH reduces harmful responses by up to 90.0% and blocks 83.3% of jailbreak attempts in the DeepSeek-R1-Distill-Llama-8B model, while requiring 295.9x less compute than Direct Refusal and 314.1x less than SafeChain. We further introduce a zero-shot variant that requires no fine-tuning. In addition, we provide a comprehensive analysis of how existing methods in LLMs generalize, or fail, when applied to reasoning-centric models, revealing critical gaps and new directions for safer AI.

arXiv:2406.00954v2 Announce Type: replace-cross Abstract: Various machine learning approaches have gained significant popularity for the automated classification of educational text to identify indicators of learning engagement -- i.e. learning engagement classification (LEC). LEC can offer comprehensive insights into human learning processes, attracting significant interest from diverse research communities, including Natural Language Processing (NLP), Learning Analytics, and Educational Data Mining. Recently, Large Language Models (LLMs), such as ChatGPT, have demonstrated remarkable performance in various NLP tasks. However, their comprehensive evaluation and improvement approaches in LEC tasks have not been thoroughly investigated. In this study, we propose the Annotation Guidelines-based Knowledge Augmentation (AGKA) approach to improve LLMs. AGKA employs GPT 4.0 to retrieve label definition knowledge from annotation guidelines, and then applies the random under-sampler to select a few typical examples. Subsequently, we conduct a systematic evaluation benchmark of LEC, which includes six LEC datasets covering behavior classification (question and urgency level), emotion classification (binary and epistemic emotion), and cognition classification (opinion and cognitive presence). The study results demonstrate that AGKA can enhance non-fine-tuned LLMs, particularly GPT 4.0 and Llama 3 70B. GPT 4.0 with AGKA few-shot outperforms full-shot fine-tuned models such as BERT and RoBERTa on simple binary classification datasets. However, GPT 4.0 lags in multi-class tasks that require a deep understanding of complex semantic information. Notably, Llama 3 70B with AGKA is a promising combination based on open-source LLM, because its performance is on par with closed-source GPT 4.0 with AGKA. In addition, LLMs struggle to distinguish between labels with similar names in multi-class classification.

arXiv:2507.00418v2 Announce Type: replace-cross Abstract: This study presents a benchmarking analysis of the Qualcomm Cloud AI 100 Ultra (QAic) accelerator for large language model (LLM) inference, evaluating its energy efficiency (throughput per watt), performance, and hardware scalability against NVIDIA A100 GPUs (in 4x and 8x configurations) within the National Research Platform (NRP) ecosystem. A total of 12 open-source LLMs, ranging from 124 million to 70 billion parameters, are served using the vLLM framework. Our analysis reveals that QAic achieves competitive energy efficiency with advantages on specific models while enabling more granular hardware allocation: some 70B models operate on as few as 1 QAic card versus 8 A100 GPUs required, with 20x lower power consumption (148W vs 2,983W). For smaller models, single QAic devices achieve up to 35x lower power consumption compared to our 4-GPU A100 configuration (36W vs 1,246W). The findings offer insights into the potential of the Qualcomm Cloud AI 100 Ultra for energy-constrained and resource-efficient HPC deployments within the National Research Platform (NRP).

arXiv:2508.04586v4 Announce Type: replace-cross Abstract: Artificial Intelligence (AI) conferences are essential for advancing research, sharing knowledge, and fostering academic community. However, their rapid expansion has rendered the centralized conference model increasingly unsustainable. This paper offers a data-driven diagnosis of a structural crisis that threatens the foundational goals of scientific dissemination, equity, and community well-being. We identify four key areas of strain: (1) scientifically, with per-author publication rates more than doubling over the past decade to over 4.5 papers annually; (2) environmentally, with the carbon footprint of a single conference exceeding the daily emissions of its host city; (3) psychologically, with 71% of online community discourse reflecting negative sentiment and 35% referencing mental health concerns; and (4) logistically, with attendance at top conferences such as NeurIPS 2024 beginning to outpace venue capacity. These pressures point to a system that is misaligned with its core mission. In response, we propose the Community-Federated Conference (CFC) model, which separates peer review, presentation, and networking into globally coordinated but locally organized components, offering a more sustainable, inclusive, and resilient path forward for AI research.

arXiv:2510.15001v2 Announce Type: replace-cross Abstract: We introduce VaultGemma 1B, a 1 billion parameter model within the Gemma family, fully trained with differential privacy. Pretrained on the identical data mixture used for the Gemma 2 series, VaultGemma 1B represents a significant step forward in privacy-preserving large language models. We openly release this model to the community

Nature Biotechnology, Published online: 23 October 2025; doi:10.1038/s41587-025-02855-x

Guidance for optimizing lipid nanoparticle formulations is derived using sophisticated biophysical techniques.