Liwei Wang
YOU?
Author Swipe
View article: NeMo: Needle in a Montage for Video-Language Understanding
NeMo: Needle in a Montage for Video-Language Understanding Open
Recent advances in video large language models (VideoLLMs) call for new evaluation protocols and benchmarks for complex temporal reasoning in video-language understanding. Inspired by the needle in a haystack test widely used by LLMs, we i…
View article: Characterization of strontium-rich groundwater in a typical alpine basin on Tibetan Plateau: Implications for sustainable exploitation and development
Characterization of strontium-rich groundwater in a typical alpine basin on Tibetan Plateau: Implications for sustainable exploitation and development Open
This study focuses on mineral groundwater in alpine regions and its sustainable exploitation. The Tongde basin on Tibetan Plateau was investigated to reveal the hydrochemistry and formation of mineral groundwater in alpine basins and its s…
View article: Context Matching is not Reasoning: Assessing Generalized Evaluation of Generative Language Models in Clinical Settings
Context Matching is not Reasoning: Assessing Generalized Evaluation of Generative Language Models in Clinical Settings Open
Current discussion surrounding the clinical capabilities of generative language models (GLMs) predominantly center around multiple-choice question-answer (MCQA) benchmarks derived from clinical licensing examinations. While accepted for hu…
View article: Hydrogeochemical signatures, genetic mechanisms, and sustainable utilization potential of Li-B-Sr enriched groundwater in a typical arid endorheic watershed on Tibetan Plateau
Hydrogeochemical signatures, genetic mechanisms, and sustainable utilization potential of Li-B-Sr enriched groundwater in a typical arid endorheic watershed on Tibetan Plateau Open
Study region: Nalenggele watershed, the largest endorheic watershed in the hyper-arid Qaidam Basin, Tibetan Plateau. Study focus: Tibetan Plateau endorheic basins exhibit significant trace element enrichment in water systems, enabling salt…
View article: Hierarchical in-out fusion for incomplete multimodal brain tumor segmentation
Hierarchical in-out fusion for incomplete multimodal brain tumor segmentation Open
Fusing multimodal data play a crucial role in accurate brain tumor segmentation network and clinical diagnosis, especially in scenarios with incomplete multimodal data. Existing multimodal fusion models usually perform intra-modal fusion a…
View article: Teaching innovation in a pharmacy course: integration of “Questioning-Training of Comprehensive Knowledge Application” and a “Teacher-AI-Student Interaction Model”
Teaching innovation in a pharmacy course: integration of “Questioning-Training of Comprehensive Knowledge Application” and a “Teacher-AI-Student Interaction Model” Open
Background In the context of the rapid development of artificial intelligence (AI) technology, questioning has become an increasingly active and critical component of effective collaboration with AI. Also, the ability to apply knowledge fr…
View article: Diagnosing and Improving Diffusion Models by Estimating the Optimal Loss Value
Diagnosing and Improving Diffusion Models by Estimating the Optimal Loss Value Open
Diffusion models have achieved remarkable success in generative modeling. Despite more stable training, the loss of diffusion models is not indicative of absolute data-fitting quality, since its optimal value is typically not zero but unkn…
View article: Spatiotemporal variation and driving forces of soil salinization in the lower reach of arid endorheic basins: Critical role of lake system and groundwater overflow
Spatiotemporal variation and driving forces of soil salinization in the lower reach of arid endorheic basins: Critical role of lake system and groundwater overflow Open
Arid endorheic basins confront the most severe soil salinization challenge globally, particularly in their lower reaches. This research focuses on the spatiotemporal dynamics and underlying mechanism of soil salinization in the lower reach…
View article: Understanding Cancer Survivorship Care Needs Using Amazon Reviews: Content Analysis, Algorithm Development, and Validation Study
Understanding Cancer Survivorship Care Needs Using Amazon Reviews: Content Analysis, Algorithm Development, and Validation Study Open
Background Complementary therapies are being increasingly used by cancer survivors. As a channel for customers to share their feelings, outcomes, and perceived knowledge about the products purchased from e-commerce platforms, Amazon consum…
View article: Deep Phenotyping of Obesity: Electronic Health Record–Based Temporal Modeling Study
Deep Phenotyping of Obesity: Electronic Health Record–Based Temporal Modeling Study Open
Background Obesity affects approximately 40% of adults and 15%‐20% of children and adolescents in the United States, and poses significant economic and psychosocial burdens. Currently, patient responses to any single antiobesity medication…
View article: Research on Amazon's Business Strategy
Research on Amazon's Business Strategy Open
Amazon, an American multinational technology company, was established in 1994 by Jeff Bezos. Its headquarters is situated in Seattle, Washington, USA. The company's business scope is extensive, encompassing areas such as e-commerce, cloud …
View article: An open natural language processing (NLP) framework for EHR-based clinical research: a case demonstration using the National COVID Cohort Collaborative (N3C)
An open natural language processing (NLP) framework for EHR-based clinical research: a case demonstration using the National COVID Cohort Collaborative (N3C) Open
Despite recent methodology advancements in clinical natural language processing (NLP), the adoption of clinical NLP models within the translational research community remains hindered by process heterogeneity and human factor variations. C…
View article: Homomorphism Expressivity of Spectral Invariant Graph Neural Networks
Homomorphism Expressivity of Spectral Invariant Graph Neural Networks Open
Graph spectra are an important class of structural features on graphs that have shown promising results in enhancing Graph Neural Networks (GNNs). Despite their widespread practical use, the theoretical understanding of the power of spectr…
View article: Impact of oat grain supplementation on growth performance, rumen microbiota, and fatty acid profiles in Hu sheep
Impact of oat grain supplementation on growth performance, rumen microbiota, and fatty acid profiles in Hu sheep Open
The intestinal microbiota plays a vital role in animal growth and development. In this study, we explored the impact of oat grain dietary supplementation on growth performance, intestinal microbiota, short-chain fatty acids (SCFAs), and fa…
View article: [Clinical application of multidisciplinary team in the diagnosis and treatment of chronic refractory wounds].
[Clinical application of multidisciplinary team in the diagnosis and treatment of chronic refractory wounds]. Open
The MDT mode can significantly reduce the number of surgeries for patients with chronic refractory wounds, improve the effectiveness of therapy and increase patient satisfaction. It is a recommended model for optimizing the clinical diagno…
View article: GraphMorph: Tubular Structure Extraction by Morphing Predicted Graphs
GraphMorph: Tubular Structure Extraction by Morphing Predicted Graphs Open
Accurately restoring topology is both challenging and crucial in tubular structure extraction tasks, such as blood vessel segmentation and road network extraction. Diverging from traditional approaches based on pixel-level classification, …
View article: Learning to Reason from Feedback at Test-Time
Learning to Reason from Feedback at Test-Time Open
Solving complex tasks in a single attempt is challenging for large language models (LLMs). Iterative interaction with the environment and feedback is often required to achieve success, making effective feedback utilization a critical topic…
View article: Understanding cancer symptom management using Amazon reviews: a human-annotated corpus (Preprint)
Understanding cancer symptom management using Amazon reviews: a human-annotated corpus (Preprint) Open
BACKGROUND Complementary therapies are being increasingly used by cancer patients. As a channel for customers to share their feelings, outcomes, ideas, and perceived knowledge about the products purchased from e-commerce platforms, Amazon…
View article: End‐to‐End Crystal Structure Prediction from Powder X‐Ray Diffraction
End‐to‐End Crystal Structure Prediction from Powder X‐Ray Diffraction Open
Powder X‐ray diffraction (PXRD) is a prevalent technique in materials characterization. While the analysis of PXRD often requires extensive human manual intervention, and most automated method only achieved at coarse‐grained level. The mor…
View article: Single-backlighter time-framed X-ray imaging diagnostics of laser plasma using a quasi-coaxial multi-channel Kirkpatrick–Baez microscope
Single-backlighter time-framed X-ray imaging diagnostics of laser plasma using a quasi-coaxial multi-channel Kirkpatrick–Baez microscope Open
The time-resolved backlight imaging of plasma is crucial for diagnosing density-dependent plasma information. It requires a high-intensity X-ray source and efficient optics. We propose a quasi-coaxial, multi-channel Kirkpatrick–Baez (KB) s…
View article: Deep phenotyping obesity using EHR data: Promise, Challenges, and Future Directions
Deep phenotyping obesity using EHR data: Promise, Challenges, and Future Directions Open
Obesity affects approximately 34% of adults and 15–20% of children and adolescents in the U.S, and poses significant economic and psychosocial burdens. Due to the multifaceted nature of obesity, currently patient responses to any single an…
View article: Current status of anti-obesity medications and performance, an EHR based survey
Current status of anti-obesity medications and performance, an EHR based survey Open
Background Over the past two decades, the Food and Drug Administration (FDA) has significantly increased the approval of anti-obesity medications (AOMs) for obesity management. Both FDA-approved AOMs (F-AOMs) and off-label AOMs (O-AOMs) ha…
View article: Geochemical fingerprints, evolution, and driving forces of groundwater in an alpine basin on Tibetan Plateau: Insights from unsupervised machine learning and objective weight allocation approaches
Geochemical fingerprints, evolution, and driving forces of groundwater in an alpine basin on Tibetan Plateau: Insights from unsupervised machine learning and objective weight allocation approaches Open
Study Region: Tongde Basin, a typical alpine basin on the northeastern Tibetan Plateau Study focus: Research on the hydrochemical pattern and availability of groundwater is very limited in alpine regions, which greatly hinders the rational…