arXiv subjects

Nan Wei

Publications and source records attributed to Nan Wei.

6 recordsLinked to original sources

arXiv2026

Visual Credit Audit for Multimodal Spatial Reasoning

Closed yes/no spatial benchmarks can reward a correct answer even when the image adds little support beyond no-image contexts. Under a fixed forced-choice interface, Visual Credit Audit (VCA) separates two estimands: whether the benchmark image gives the model's declared decision more support than text-only and blank controls, and whether the model responds to relation-specific visual evidence. The first audit is training- and label-free and does not require an answer flip. Applying labels yields dependence-credited correctness (D-CC); on correct items, it equals same-control gold-aligned positive gain, while prediction alignment extends the audit to errors. Across four open MLLMs and two spatial benchmarks, 12.73-26.25% of decisions are correct yet uncredited. Matched same-split image permutation reduces D-CC by 21.25-47.80 points, with every paired 95% interval above zero. Fixed-pixel relation contrasts and a 3x3 evidence-source factorial show why null controls cannot identify relation response. Among controlled correct-but-uncredited agreement decisions, response to relation reversal spans 81.57-100.00%, while 32.11% pooled change answer. Independently audited outcomes on 108 geometry-compatible edits provide a bounded natural-image correspondence check. VCA thereby decomposes benchmark success into correctness, additional image support, and relation-consistent response.

cs.CV↗

arXiv2026

Exascale Hybrid Numerical-AI Ensembles for Operational Flood-Season Forecasting in East Asia: 15-km Decadal Hindcasts and 1-km High-Resolution Capability

Seasonal forecasting of summer rainfall in East Asia remains a grand challenge, as predictability at 3 to 6 month lead times is constrained by the spring predictability barrier, weak large-scale signals, and localized nonlinear convective extremes. We address this challenge with CAPES, which integrates a kilometer-resolution coupled regional model with atmosphere, land, and ocean components and a data-driven AI seasonal forecasting system. At 15 km resolution, the fused workflow combines 174 numerical members from varying start times, physics schemes, and parameter perturbations with 1,600 AI members generated from initial and physical perturbations. Using the full LineShine system, CAPES completes ten annual 1,774-member hindcasts for 2016 to 2025 within 14.6 hours, improving the mean prediction score from ECMWF's 71.8 to 75.9 and delivering a major gain in operational forecasting capability. The 1-km configuration further enables fine-scale typhoon simulation and establishes the feasibility of kilometer-scale fused ensemble forecasting on a one-week timescale.

cs.CE↗

arXiv2019

Evaluation of soil thermal conductivity schemes for use in land surface modeling

Soil thermal conductivity is an important physical parameter in modeling land surface processes. Previous studies on evaluations of parameterization schemes of soil thermal conductivity are mostly based on specific experimental conditions or local soil samples, and their recommendations may not be the optimal schemes for land surface model (LSMs). In this work, seven highly recommended soil thermal conductivity schemes are evaluated for their applicability in LSMs. With the consideration of both scheme estimations and land process simulations by incorporation into the Common Land Model, the Balland and Arp [2005] scheme is found to consistently perform best among all the schemes, and thus can be recommended as a superior scheme for land modeling use. Uncertainty analyses by in-situ simulations demonstrate that, over relatively dry regions, the inter-scheme variations of soil thermal conductivity can lead to significant differences of simulated soil temperature, especially at deep layers, due to changes of downward soil heat conduction and the associated freeze-thaw cycles. However, few effects appear over wet regions, likely due to the high soil heat capacity induced by high soil moisture levels, which increases the heat inertia in soil thermodynamics. Global comparisons show the similar relationships that soil thermal conductivity significantly affects the simulated soil temperature and other related thermal and hydraulic variables over arid and semi-arid regions in mid- and high-latitudes. These results display the role of soil thermal conductivity in LSM, and suggest the importance of the evaluation and further development of thermal conductivity schemes with respect to land modelling applications.

physics.ao-ph↗

arXiv2019

Water Preservation in Soan River Basin using Deep Learning Techniques

Water supplies are crucial for the development of living beings. However, change in the hydrological process i.e. climate and land usage are the key issues. Sustaining water level and accurate estimating for dynamic conditions is a critical job for hydrologists, but predicting hydrological extremes is an open issue. In this paper, we proposed two deep learning techniques and three machine learning algorithms to predict stream flow, given the present climate conditions. The results showed that the Recurrent Neural Network (RNN) or Long Short-term Memory (LSTM), an artificial neural network based method, outperform other conventional and machine-learning algorithms for predicting stream flow. Furthermore, we analyzed that stream flow is directly affected by precipitation, land usage, and temperature. These indexes are critical, which can be used by hydrologists to identify the potential for stream flow. We make the dataset publicly available (https://github.com/sadaqat007/Dataset) so that others should be able to replicate and build upon the results published.

cs.NE↗

arXiv2019

Gas-phase synthesis of carbon nanotube-graphene heterostructures

Graphene and carbon nanotubes (CNTs) share the same atomic structure of hexagonal carbon lattice. Yet, their synthesis differs in many aspects, including the shape and size of the catalyst. Here, we demonstrate a floating-catalyst chemical vapor deposition (FCCVD) technique for substrate-free, single-step growth of CNT-graphene heterostructures (CNT-G-H) using ethylene as a carbon source. The formation of CNT-G-H is directly evidenced by lattice-resolved (scanning) transmission electron microscopy (STEM) and electron diffraction experiments, corroborated by atomic force microscopy (AFM). Our experiments show the relative number density of graphene-nanoflakes can be tuned by optimizing the synthesis conditions. Since in the applied process the formation of the structures take place in gas-suspension, the as-synthesized CNT-G-H films can be deposited on any surface in ambient temperature with an arbitrary thickness. Moreover, this process of CNT-G-H synthesis with strong universality has also been realized in multiple systems of ethylene-based FCCVD with various catalysts and set-ups.

physics.app-ph↗

arXiv2018

TS-CNN: Text Steganalysis from Semantic Space Based on Convolutional Neural Network

Steganalysis has been an important research topic in cybersecurity that helps to identify covert attacks in public network. With the rapid development of natural language processing technology in the past two years, coverless steganography has been greatly developed. Previous text steganalysis methods have shown unsatisfactory results on this new steganography technique and remain an unsolved challenge. Different from all previous text steganalysis methods, in this paper, we propose a text steganalysis method(TS-CNN) based on semantic analysis, which uses convolutional neural network(CNN) to extract high-level semantic features of texts, and finds the subtle distribution differences in the semantic space before and after embedding the secret information. To train and test the proposed model, we collected and released a large text steganalysis(CT-Steg) dataset, which contains a total number of 216,000 texts with various lengths and various embedding rates. Experimental results show that the proposed model can achieve nearly 100\% precision and recall, outperforms all the previous methods. Furthermore, the proposed model can even estimate the capacity of the hidden information inside. These results strongly support that using the subtle changes in the semantic space before and after embedding the secret information to conduct text steganalysis is feasible and effective.

cs.CR↗