Effective urban planning is crucial for enhancing residents' quality of life and ensuring societal stability, playing a pivotal role in the sustainable development of cities. Current planning methods heavily rely on human experts, which are time-consuming and labor-intensive, or utilize deep learning algorithms, often limiting stakeholder involvement. To bridge these gaps, we propose Intelli-Planner, a novel framework integrating Deep Reinforcement Learning (DRL) with large language models (LLMs) to facilitate participatory and customized planning scheme generation. Intelli-Planner utilizes demographic, geographic data, and planning preferences to determine high-level planning requirements and demands for each functional type. During training, a knowledge enhancement module is employed to enhance the decision-making capability of the policy network. Additionally, we establish a multi-dimensional evaluation system and employ LLM-based stakeholders for satisfaction scoring. Experimental validation across diverse urban settings shows that Intelli-Planner surpasses traditional baselines and achieves comparable performance to state-of-the-art DRL-based methods in objective metrics, while enhancing stakeholder satisfaction and convergence speed. These findings underscore the effectiveness and superiority of our framework, highlighting the potential for integrating the latest advancements in LLMs with DRL approaches to revolutionize tasks related to functional areas planning. Code and data are available at https://github.com/chicosirius/Intelli-Planner.
论文检索
输入标题、作者或关键词，从 954 篇学术成果中精准定位
Mind the Ambiguity: Aleatoric Uncertainty Quantification in LLMs for Safe Medical Question Answering
The deployment of Large Language Models in Medical Question Answering is severely hampered by ambiguous user queries, a significant safety risk that demonstrably reduces answer accuracy in high-stakes healthcare settings. In this paper, we formalize this challenge by linking input ambiguity to aleatoric uncertainty (AU), which is the irreducible uncertainty arising from underspecified input. To facilitate research in this direction, we construct CV-MedBench, the first benchmark designed for studying input ambiguity in Medical QA. Using this benchmark, we analyze AU from a representation engineering perspective, revealing that AU is linearly encoded in LLM's internal activation patterns. Leveraging this insight, we introduce a novel AU-guided ''Clarify-Before-Answer'' framework, which incorporates AU-Probe - a lightweight module that detects input ambiguity directly from hidden states. Unlike existing uncertainty estimation methods, AU-Probe requires neither LLM fine-tuning nor multiple forward passes, enabling an efficient mechanism to proactively request user clarification and significantly enhance safety. Extensive experiments across four open LLMs demonstrate the effectiveness of our QA framework, with an average accuracy improvement of 9.48% over baselines. Our framework provides an efficient and robust solution for safe Medical QA, strengthening the reliability of health-related applications. The code is available at https://github.com/yaokunliu/AU-Med.git, and the CV-MedBench dataset is released on Hugging Face at https://huggingface.co/datasets/yaokunl/CV-MedBench.
Per- and polyfluoroalkyl substances (PFAS) are persistent environmental contaminants that require integrated, semantically structured representations of chemical identity, classification, and properties to support integrated contaminant monitoring and analysis. This work presents the CompTox ontology, an expert-guided ontology describing commonly analyzed PFAS and designed to support PFAS data integration and querying. The ontology organizes PFAS hierarchically according to key physicochemical characteristics and incorporates authoritative identifiers and properties from EPA's CompTox Chemicals Dashboard. Individual PFAS are annotated with core chemical identifiers, including DTXSID, CASRN, InChIKey, and SMILES; with physicochemical attributes such as molecular mass, carbon chain length, and functional group information; and with observed or predicted environmental fate and transport and toxicological information. The ontology was constructed using the Knowledge Acquisition and Representation Methodology (KNARM), employing a template-driven workflow implemented with the ROBOT tool to generate an OWL-formatted ontology. An expert-guided hierarchy captures major PFAS classes, including fluorotelomers, perfluoroalkyl acids (both Perfluoroalkyl Carboxylic and Sulfonic Acids), and perfluoroalkyl ether acids. Human-readable IRIs and SKOS alternative labels enhance usability. The ontology helps facilitate integrated querying and analysis of PFAS contamination within the SAWGraph knowledge graphs but also serves as a flexible and extensible framework for unified chemical identification and classification.
The escalating energy consumption and carbon footprint of training large-scale Web AI models pose urgent challenges for sustainable development. Dataset Distillation (DD) offers a promising avenue for green AI by compressing large datasets into small synthetic ones for efficient training. However, most existing DD methods overfit to the inductive biases of specific source architectures (e.g., CNNs or ViTs), resulting in poor cross-model generalization. This limitation necessitates redundant re-distillation processes for different architectures, severely undermining the energy-saving potential of DD. To address this, we introduce Adversarial Mutual Information Distillation (AMID), a rigorous framework designed to create highly reusable and robust synthetic datasets. From an information-theoretic perspective, we cast model-agnosticism as minimizing the mutual information (MI) between the synthetic data and the specific identity of the distillation model. We convert this intractable objective into a tractable two-player adversarial game, which unifies knowledge preservation with adversarial unlearning of architectural bias. Extensive experiments on CIFAR-10 and Tiny ImageNet demonstrate that AMID achieves state-of-the-art cross-architecture generalization across diverse CNNs and ViTs. Crucially, our analysis confirms that AMID significantly reduces the computational overhead and CO2 emissions of downstream training while maintaining robust performance, paving the way for energy-efficient, transferable, and sustainable Web AI ecosystems.
Web-based platforms are becoming a primary channel for psychological support, yet most LLM-driven chatbots remain opaque, single-stage, and weakly grounded in established therapeutic practice. To address this gap, we present XInsight, a multi-agent framework that models psychological support as a stage-consistent workflow aligned with the classical Exploration-Insight-Action paradigm. Building on structured client representations, XInsight orchestrates specialized agents under a unified Reason-Intervene-Reflect cycle: an Exploration agent organizes background and concerns into a structured Case Conceptualization Form, a Routing agent performs Adaptive Therapeutic Routing (ATR) across SFBT, CBT, and MBCT, a unified Therapeutic agent executes school-consistent submodules, and a Consolidation agent guides review, skill integration, and relapse-prevention planning. A Recording agent continuously transforms open-ended web dialogues into standardized psychological artifacts, enhancing interpretability, continuity, and accountability. To support transparent assessment, we introduce XInsight-Bench with a Scale-Guided LLM Evaluation (SGLE) protocol that combines therapy-specific clinical scales with general counseling criteria. Experiments show improved paradigm alignment, multi-therapy integration, interaction depth, and interpretability over existing multi-agent counseling systems, indicating that XInsight provides a practical blueprint for integrating counseling-inspired support agents into web applications for digital well-being.
Large Language Models (LLMs) are increasingly used for toxicity assessment in online moderation systems, where fairness across demographic groups is essential for equitable treatment. However, LLMs often produce inconsistent toxicity judgements for subtle expressions, particularly those involving implicit hate speech, revealing underlying biases that are difficult to correct through standard training. This raises a key question that existing approaches often overlook: when should corrective mechanisms be invoked to ensure fair and reliable assessments? To address this, we propose FairToT, an inference-time framework that enhances LLM fairness through prompt-guided toxicity assessment. FairToT identifies cases where demographic-related variation is likely to occur and determines when additional assessment should be applied. In addition, we introduce two interpretable fairness indicators that detect such cases and improve inference consistency without modifying model parameters. Experiments on benchmark datasets show that FairToT reduces group-level disparities while maintaining stable and reliable toxicity predictions, demonstrating that inference-time refinement offers an effective and practical approach for fairness improvement in LLM-based toxicity assessment systems. The source code can be found at https://aisuko.github.io/fair-tot/.
Precision pathology relies on detecting fine-grained morphological abnormalities within specific Regions of Interest (ROIs), as these local, texture-rich cues—rather than global slide contexts—drive expert diagnostic reasoning. While Vision-Language (V-L) models promise data efficiency by leveraging semantic priors, adapting them faces a critical Granularity Mismatch, where generic representations fail to resolve such subtle defects. Current adaptation methods often treat modalities as independent streams, failing to ground semantic prompts in ROI-specific visual contexts. To bridge this gap, we propose the Hierarchical Adaptation and Alignment Framework (HAAF ). At its core is a novel Cross-Level Scaled Alignment (CLSA) mechanism that enforces a sequential calibration order: visual features first inject context into text prompts to generate content-adaptive descriptors, which then spatially guide the visual encoder to spotlight anomalies. Additionally, a dual-branch inference strategy integrates semantic scores with geometric prototypes to ensure stability in few-shot settings. Experiments on four benchmarks show HAAF significantly outperforms state-of-the-art methods and effectively scales with domain-specific backbones (e.g., CONCH) in low-resource scenarios.
As web platforms evolve towards greater personalization and emotional complexity, conversational agents must transcend superficial empathy to demonstrate identity-aware emotional reasoning. However, existing systems face two limitations: (1) reliance on situation-centric datasets lacking persistent user identity, which hampers the capture of personalized affective nuances; and (2) dependence on opaque, coarse reward signals that hinder development of verifiable empathetic reasoning. To address these gaps, we introduce KardiaBench, a large-scale user-grounded benchmark comprising 178,080 QA pairs across 22,080 multi-turn conversations anchored to 671 real-world profiles. The dataset is constructed via a model-in-the-loop pipeline with iterative rubric-guided refinement to ensure psychological plausibility and persona consistency. This progressive empathy pipeline that integrates user comprehension, contextual reasoning, and emotion perception into conversations, followed by iterative critique and rubric-based refinement to ensure psychological plausibility, emotional fidelity, and persona consistency. Building on this, we propose Kardia-R1, a framework that trains models for interpretable, stepwise empathetic cognition. Kardia-R1 leverages Rubric-as-Judge Empathetic Reinforcement Learning (Rubric-ERL), a GRPO-based method that uses explainable, human-aligned rubric rewards to tightly couple user understanding, emotional inference, and supportive response generation. Extensive experiments across four LLM backbones demonstrate that Kardia-R1 consistently outperforms other methods in emotion accuracy, empathy, relevance, persona consistency, and safety.
In remote or underserved regions, where road networks are often either unavailable or poorly mapped, and GPS signals are sparse and unreliable, the quality of individual trajectories is severely compromised. In such contexts, web crowdsourced data becomes essential for accurately recovering trajectories and compensating for missing spatial information in the absence of explicit road networks. However, this situation introduces two key challenges:(i) scarcity of data in unseen regions, restricting transfer learning; and (ii) the necessity to infer latent movement structures under roadless conditions. To address these, we propose Region-aware Hierarchical Trajectory Recovery (RHTR) model, designed for location inference from web crowdsourced data in sparse, roadless scenarios. RHTR constructs multi-scale implicit grid-based offset maps from historical location data, with coarse grids capturing global patterns and fine grids refining local details. These sequential representations form region-aware encodings by sampling information around observed points, facilitating trajectory recovery. A coarse-to-fine mechanism leverages contextual information to progressively reconstruct missing segments. Experiments on two public datasets—simulating underserved and disaster-like settings with cross-region transfer—demonstrate that RHTR achieves state-of-the-art performance in trajectory recovery.
Energy consumption has become a bottleneck for future computing architectures, from wearable devices to leadership-class supercomputers. Existing energy management techniques largely target CPUs, even though GPUs now dominate power draw in heterogeneous high performance computing (HPC) systems. Moreover, many prior methods rely on either purely offline or hybrid offline and online training, which is impractical and results in energy inefficiencies during data collection. In this paper, we introduce a practical online GPU energy optimization problem in a HPC scenarios. The problem is challenging because (1) GPU frequency scaling exhibits performance–energy trade-offs, (2) online control must balance exploration and exploitation, and (3) frequent frequency switching incurs non-trivial overhead and degrades quality of service (QoS). To address the challenges, we formulate online GPU energy optimization as a multi-armed bandit problem and propose EnergyUCB, a lightweight UCB-based controller that dynamically adjusts GPU core frequency in real time to save energy. Specifically, EnergyUCB (1) defines a reward that jointly captures energy and performance using a core-to-uncore utilization ratio as a proxy for GPU throughput, (2) employs optimistic initialization and UCB-style confidence bonuses to accelerate learning from scratch, and (3) incorporates a switching-aware UCB index and a QoS-constrained variant that enforce explicit slowdown budgets while discouraging unnecessary frequency oscillations. Extensive experiments on real-world workloads from the world's third fastest supercomputer Aurora show that EnergyUCB achieves substantial energy savings with modest slowdown and that the QoS-constrained variant reliably respects user-specified performance budgets.
The rapid proliferation of emotion-aware web services has necessitated the analysis of multimodal user interactions. However, this introduces new vulnerabilities where adversaries exploit emotional signals to circumvent fraud detection systems. Despite its improved utility, the robustness of multimodal fraud detection against emotion-driven adversarial manipulation remains significantly underexplored. Existing paradigms often treat emotional cues as static features, overlooking the adversary's capability to strategically modulate multimodal signals (e.g., facial micro-expressions, vocal intonation, and textual styles) to mimic genuine behavior. Furthermore, prevalent evaluations are typically confined to unimodal perturbations and fail to account for context-consistent, cross-modal attacks, thereby compromising system reliability in real-world deployments. To bridge this gap, we propose Context-Emotion Adversarial Training (CEAT), a robust framework designed to fortify multimodal fraud detection against emotion-based attacks. CEAT leverages a Transformer-based architecture to synergistically model emotional features (e.g., visual dynamics and acoustic prosody) alongside semantic context derived from text, yielding a unified representation. Crucially, CEAT introduces a context-aware perturbation mechanism that injects noise into the emotional latent space during training. This process preserves semantic consistency while encouraging the learning of emotion-invariant and discriminative representations. Additionally, a contrastive learning objective is integrated to maximize the distributional divergence between genuine and adversarial samples within the latent manifold. Extensive experiments on multimodal benchmarks demonstrate that CEAT significantly outperforms state-of-the-art baselines, exhibiting superior robustness under simulated emotion-driven attack scenarios.
The rapid proliferation of VTubers --- digital avatars controlled and voiced by human actors (Nakanohito) --- has created a lucrative and popular entertainment ecosystem. However, the prevailing industry model, where corporations retain ownership of the VTuber persona while the Nakanohito bears the immense pressure of dual-identity management, exposes the Nakanohito to significant vulnerabilities, including burnout, harassment, and precarious labor conditions. When these pressures become untenable, the Nakanohito may terminate their contracts and later debut with a new persona, a process known as ''reincarnation''. This phenomenon, a rising concern in the industry, inflicts substantial losses on the Nakanohito, agencies, and audiences alike. Understanding the quantitative fallout of reincarnation is crucial for mitigating this damage and fostering a more sustainable industry. To address this gap, we conduct the first large-scale empirical study of VTuber reincarnation, analyzing 12 significant cases using a comprehensive dataset of 728K livestream sessions and 4.5B viewer interaction records. Our results suggest reincarnation significantly damages a Nakanohito's career, leading to a decline in audience and financial support, an increase in harassment, and negative repercussions for the wider VTuber industry. Overall, these insights carry immediate implications for mitigating the significant professional and personal costs of the reincarnation, and fostering a healthier and more equitable VTuber ecosystem.