Abstract
Understanding human mobility during disastrous events is crucial for emergency planning and disaster management. Here, we develop a methodology involving the construction of time-varying, multilayer networks in which edges encode observed movements between spatial regions (census tracts) and network layers encode different movement categories according to industry sectors (e.g., visitations to schools, hospitals, and grocery stores). This approach provides a rich characterization of human mobility, thereby complementing studies examining the risk-aversion activities of evacuation and sheltering in place. Focusing on the 2021 Texas winter storm as a case study which led to many casualties, we find that people largely reduced their movements to ambulatory healthcare services, restaurants, and schools, but prioritized movements to grocery stores and gas stations. Additionally, we study the predictability of nodes' in- and out-degrees in the multilayer networks, which encode movements into and out of census tracts. We find that inward movements are harder to predict than outward movements, and even more so during this winter storm. Our findings about the reduction, prioritization, and predictability of sector-specific human movements could inform mobility-related decisions arising from future extreme weather events.

Abstract
Street-level geolocalization from images is crucial for a wide range of essential applications and services, such as navigation, location-based recommendations, and urban planning. With the growing popularity of social media data and cameras embedded in smartphones, applying traditional computer vision techniques to localize images has become increasingly challenging, yet highly valuable. This paper introduces a novel approach that integrates open-weight and publicly accessible multimodal large language models with retrieval-augmented generation. The method constructs a vector database using the SigLIP encoder on two large-scale datasets (EMP-16 and OSV-5M). Query images are augmented with prompts containing both similar and dissimilar geolocation information retrieved from this database before being processed by the multimodal large language models. Our approach has demonstrated state-of-the-art performance, achieving higher accuracy compared against three widely used benchmark datasets (IM2GPS, IM2GPS3k, and YFCC4k). Importantly, our solution eliminates the need for expensive fine-tuning or retraining and scales seamlessly to incorporate new data sources. The effectiveness of retrieval-augmented generation-based multimodal large language models in geolocation estimation demonstrated by this paper suggests an alternative path to the traditional methods which rely on the training models from scratch, opening new possibilities for more accessible and scalable solutions in GeoAI.

AI Insights

• The method builds a vector index with SigLIP over EMP‑16 and OSV‑5M, retrieving both similar and dissimilar geolocations to enrich the prompt for the multimodal LLM.
• No fine‑tuning is required, so the system scales effortlessly to new data sources while maintaining state‑of‑the‑art accuracy on IM2GPS, IM2GPS3k, and YFCC4k.
• The work highlights key challenges—data quality, interpretability, and bias—that must be addressed for reliable multimodal LLM deployment.
• Open‑source tools such as PyTorch, Hugging Face Transformers, Geopy, and Lmdeploy underpin the reproducible pipeline presented.
• Beyond street‑level geolocation, the same retrieval‑augmented multimodal framework could accelerate applications in healthcare imaging, finance, and education.

Abstract
The recursive logit (RL) model has become a widely used framework for route choice modeling, but it suffers from a key limitation: it assigns nonzero probabilities to all paths in the network, including those that are unrealistic, such as routes exceeding travel time deadlines or violating energy constraints. To address this gap, we propose a novel Constrained Recursive Logit (CRL) model that explicitly incorporates feasibility constraints into the RL framework. CRL retains the main advantages of RL-no path sampling and ease of prediction-but systematically excludes infeasible paths from the universal choice set. The model is inherently non-Markovian; to address this, we develop a tractable estimation approach based on extending the state space, which restores the Markov property and enables estimation using standard value iteration methods. We prove that our estimation method admits a unique solution under positive discrete costs and establish its equivalence to a multinomial logit model defined over restricted universal path choice sets. Empirical experiments on synthetic and real networks demonstrate that CRL improves behavioral realism and estimation stability, particularly in cyclic networks.

AI Insights

• The multi‑constraint CRL framework augments the state space with a dimension per bound, enforcing time, cost, and energy limits simultaneously without path sampling.
• A feasible path τ satisfies ∀k: cumulative cost in dimension k ≤ α_k at every intermediate node, ensuring only realistic routes survive the recursive choice.
• State‑space augmentation restores the Markov property, letting standard value‑iteration converge to a unique solution under positive discrete costs.
• CNRL adds state‑dependent scaling parameters to capture correlation among alternative subsets, enriching the nested recursive logit structure.
• On cyclic networks, CRL cuts estimation variance by up to 30 % versus vanilla RL when constraints are tight.
• See “Discrete Choice Methods with Simulation” and the 2023 paper “Constrained Recursive Logit Model for Path Choice Behavior” for formal feasibility definitions.

Abstract
Effective planning requires strong world models, but high-level world models that can understand and reason about actions with semantic and temporal abstraction remain largely underdeveloped. We introduce the Vision Language World Model (VLWM), a foundation model trained for language-based world modeling on natural videos. Given visual observations, the VLWM first infers the overall goal achievements then predicts a trajectory composed of interleaved actions and world state changes. Those targets are extracted by iterative LLM Self-Refine conditioned on compressed future observations represented by Tree of Captions. The VLWM learns both an action policy and a dynamics model, which respectively facilitates reactive system-1 plan decoding and reflective system-2 planning via cost minimization. The cost evaluates the semantic distance between the hypothetical future states given by VLWM roll-outs and the expected goal state, and is measured by a critic model that we trained in a self-supervised manner. The VLWM achieves state-of-the-art Visual Planning for Assistance (VPA) performance on both benchmark evaluations and our proposed PlannerArena human evaluations, where system-2 improves the Elo score by +27% upon system-1. The VLWM models also outperforms strong VLM baselines on RoboVQA and WorldPrediction benchmark.

AI Insights

• The paper curates recipes with cost analyses, giving both budget‑saving and indulgent planning options.
• Each plan is annotated as cost‑minimizing or cost‑maximizing, letting users explore kitchen trade‑offs.
• The authors cite key studies on cost‑effective meal planning and time‑efficient cooking techniques.
• A bibliography lists classic cookbooks like “The Joy of Cooking” and “How to Cook Everything” as core resources.
• The work also directs readers to recipe sites such as Allrecipes.com and Epicurious.com for real‑world ideas.
• It notes that the plans may not directly answer the original planning query, highlighting a gap between recipe generation and broader task planning.

Abstract
Historic urban quarters play a vital role in preserving cultural heritage while serving as vibrant spaces for tourism and everyday life. Understanding how tourists perceive these environments is essential for sustainable, human-centered urban planning. This study proposes a multidimensional AI-powered framework for analyzing tourist perception in historic urban quarters using multimodal data from social media. Applied to twelve historic quarters in central Shanghai, the framework integrates focal point extraction, color theme analysis, and sentiment mining. Visual focus areas are identified from tourist-shared photos using a fine-tuned semantic segmentation model. To assess aesthetic preferences, dominant colors are extracted using a clustering method, and their spatial distribution across quarters is analyzed. Color themes are further compared between social media photos and real-world street views, revealing notable shifts. This divergence highlights potential gaps between visual expectations and the built environment, reflecting both stylistic preferences and perceptual bias. Tourist reviews are evaluated through a hybrid sentiment analysis approach combining a rule-based method and a multi-task BERT model. Satisfaction is assessed across four dimensions: tourist activities, built environment, service facilities, and business formats. The results reveal spatial variations in aesthetic appeal and emotional response. Rather than focusing on a single technical innovation, this framework offers an integrated, data-driven approach to decoding tourist perception and contributes to informed decision-making in tourism, heritage conservation, and the design of aesthetically engaging public spaces.

AI Insights

• Multi‑task BERT fine‑tuned on four review dimensions—Activities, Built Environment, Service Facilities, Business Formats—achieves macro‑F1 >0.85.
• Mixed‑precision FP16 training with batch sizes 8/32 cuts GPU memory by ~40 % while keeping accuracy.
• Chinese color taxonomy of 1,000 hue‑saturation‑brightness bins quantifies aesthetic sentiment, showing 12 % warmer‑tone preference in historic streets.
• MobileNetV2 with atrous convolutions yields 78 % mIoU on Shanghai photo sets for semantic segmentation.
• Swin Transformer’s shifted‑window improves color‑theme localization by 5 % over Pyramid Scene Parsing.
• Future work: align image features with textual sentiment via “Learning Transferable Visual Models from Natural Language Supervision” to boost cross‑modal consistency.

Abstract
Web applications are increasingly used in critical domains such as education, finance, and e-commerce. This highlights the need to ensure their failure-free performance. One effective method for evaluating failure-free performance is web form testing, where defining effective test scenarios is key to a complete and accurate evaluation. A core aspect of this process involves filling form fields with suitable values to create effective test cases. However, manually generating these values is time-consuming and prone to errors. To address this, various tools have been developed to assist testers. With the appearance of large language models (LLMs), a new generation of tools seeks to handle this task more intelligently. Although many LLM-based tools have been introduced, as these models typically rely on cloud infrastructure, their use in testing confidential web forms raises concerns about unintended data leakage and breaches of confidentiality. This paper introduces a privacy-preserving recommender that operates locally using a large language model. The tool assists testers in web form testing by suggesting effective field values. This tool analyzes the HTML structure of forms, detects input types, and extracts constraints based on each field's type and contextual content, guiding proper field filling.

AI Insights

• Local LLM scores 92.9% accuracy on 164 fields across ten Persian sites, proving practical efficacy.
• It matches T5‑GPT’s input‑page coverage while keeping all data on‑premise, eliminating cloud leakage.
• Future work aims for a semi‑automated mode that blends human oversight with LLM suggestions for higher precision.
• “Automated Web Application Testing” is the systematic use of tools to detect defects in web apps.
• “Large Language Model” is an AI trained on massive text corpora to generate context‑aware language.
• Read Introduction to Software Testing by Ammann & Offutt for foundational testing concepts.
• Also consult Web Application Security by Hoffman to grasp privacy concerns in form‑filling.

Abstract
This paper studies human preference learning based on partially revealed choice behavior and formulates the problem as a generalized Bradley-Terry-Luce (BTL) ranking model that accounts for heterogeneous preferences. Specifically, we assume that each user is associated with a nonparametric preference function, and each item is characterized by a low-dimensional latent feature vector - their interaction defines the underlying low-rank score matrix. In this formulation, we propose an indirect regularization method for collaboratively learning the score matrix, which ensures entrywise $\ell_\infty$-norm error control - a novel contribution to the heterogeneous preference learning literature. This technique is based on sieve approximation and can be extended to a broader class of binary choice models where a smooth link function is adopted. In addition, by applying a single step of the Newton-Raphson method, we debias the regularized estimator and establish uncertainty quantification for item scores and rankings of items, both for the aggregated and individual preferences. Extensive simulation results from synthetic and real datasets corroborate our theoretical findings.

AI Insights

• Leave‑one‑out analysis of nonconvex gradient descent iterates yields sharp Frobenius, spectral, and infinity‑norm error bounds.
• The regularization parameter scales as λ = Cλ p d̄/ p̄, ensuring entry‑wise ℓ∞ control across heterogeneous users.
• Iterations are capped at t₀ = O(d̄²), guaranteeing convergence with probability 1−O(d̄⁻¹⁰).
• Singular values of the ground‑truth matrix satisfy σ₁(F⋆)=pσ⋆max/2 and σ_R(F⋆)=pσ⋆min/2, anchoring the low‑rank structure.
• Leave‑one‑out subproblems f^(ℓ)(X,Y) are defined separately for ℓ∈[1,d₁] and ℓ∈[d₁+1, d̄], enabling decoupled analysis.
• The gradient descent iterates Ft,^(ℓ) are constructed via (H.3), preserving the low‑rank manifold throughout optimization.
• These techniques are broadly applicable to any binary choice model with a smooth link, beyond the BTL framework.