Jiaxing Huang
YOU?
Author Swipe
View article: Towards Efficient Multimodal Unified Reasoning Model via Model Merging
Towards Efficient Multimodal Unified Reasoning Model via Model Merging Open
Although Multimodal Large Language Models (MLLMs) have demonstrated remarkable capabilities across diverse tasks, they encounter challenges in terms of reasoning efficiency, large model size and overthinking. However, existing lightweight …
View article: Versatile phenolic composites by in situ polymerization of concentrated dispersions of carbon nanotubes
Versatile phenolic composites by in situ polymerization of concentrated dispersions of carbon nanotubes Open
Uniform dispersion of carbon nanotubes in a polymer matrix is a prerequisite for high-performance nanotube-based composites. Here, we report an in situ polymerization route to synthesize a range of phenolic composites with high loading of …
View article: Molecular Dynamics Insights into Bio-Oil-Enhanced Self-Healing of Aged Asphalt
Molecular Dynamics Insights into Bio-Oil-Enhanced Self-Healing of Aged Asphalt Open
Long-term aging deteriorates asphalt’s self-healing capacity, yet the molecular mechanisms of bio-oil rejuvenation remain unclear. The fluidity and healing index of an asphalt binder were tested using a dynamic shear rheometer, and a heali…
View article: A method of rice panicle number counting based on improved CSRNet model
A method of rice panicle number counting based on improved CSRNet model Open
Introduction Rice is one of the world's leading food crops, with nearly half of the world's population eating rice as their staple food. Rice yield is directly related to varieties, and the most intuitive agronomic trait of varietal yield …
View article: Training-Free Text-Guided Image Editing with Visual Autoregressive Model
Training-Free Text-Guided Image Editing with Visual Autoregressive Model Open
Text-guided image editing is an essential task that enables users to modify images through natural language descriptions. Recent advances in diffusion models and rectified flows have significantly improved editing quality, primarily relyin…
View article: R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimization
R1-VL: Learning to Reason with Multimodal Large Language Models via Step-wise Group Relative Policy Optimization Open
Recent studies generally enhance MLLMs' reasoning capabilities via supervised fine-tuning on high-quality chain-of-thought reasoning data, which often leads models to merely imitate successful reasoning paths without understanding what the…
View article: Passivated Zn Powders as Metal Anode
Passivated Zn Powders as Metal Anode Open
Impacted by heavy corrosion and poor connections, zinc (Zn) powders have rarely been considered as the raw materials of Zn‐ion aqueous batteries (ZABs). Nonetheless, the ease of controlling loadings of Zn powders entitles ZABs to better ca…
View article: Discovery of Novel Low Bee‐Toxicity Aphicidal <i>α</i>‐Butenolide: Design and Optimization Based on a Progressive Strategy of Acetylcholine Binding Protein and Calculated Oil–Water Partition Coefficient
Discovery of Novel Low Bee‐Toxicity Aphicidal <i>α</i>‐Butenolide: Design and Optimization Based on a Progressive Strategy of Acetylcholine Binding Protein and Calculated Oil–Water Partition Coefficient Open
The development of novel insecticides with low bee toxicity has become increasingly urgent as many high bee‐toxicity neonicotinoids have been progressively restricted. In this study, novel halogenated phenyl‐substituted α ‐butenolide compo…
View article: Learnable Evolutionary Multi-Objective Combinatorial Optimization via Sequence-to-Sequence Model
Learnable Evolutionary Multi-Objective Combinatorial Optimization via Sequence-to-Sequence Model Open
Recent advances in learnable evolutionary algorithms have demonstrated the importance of leveraging population distribution information and historical evolutionary trajectories. While significant progress has been made in continuous optimi…
View article: Ascorbic acid is involved in melatonin-induced salinity tolerance of maize (Zea mays L.) by regulating antioxidant and photosynthetic capacities
Ascorbic acid is involved in melatonin-induced salinity tolerance of maize (Zea mays L.) by regulating antioxidant and photosynthetic capacities Open
Melatonin (MT), an indole compound, can boost plant growth under abiotic stress conditions. This experiment aims to elucidate the synergistic effect of MT and ascorbic acid (AsA) in mitigating salinity stress by assessing the photosyntheti…
View article: SPAgent: Adaptive Task Decomposition and Model Selection for General Video Generation and Editing
SPAgent: Adaptive Task Decomposition and Model Selection for General Video Generation and Editing Open
While open-source video generation and editing models have made significant progress, individual models are typically limited to specific tasks, failing to meet the diverse needs of users. Effectively coordinating these models can unlock a…
View article: A Survey on Vision Autoregressive Model
A Survey on Vision Autoregressive Model Open
Autoregressive models have demonstrated great performance in natural language processing (NLP) with impressive scalability, adaptability and generalizability. Inspired by their notable success in NLP field, autoregressive models have been …
View article: Robust hybrid diffusion control for long-term scalable frost prevention
Robust hybrid diffusion control for long-term scalable frost prevention Open
Antifrosting surfaces are critical to the efficient and safe operation of infrastructure in cold and humid environments where deposition of frost (porous ice) is thermodynamically inevitable. Such infrastructure can include above-ground po…
View article: Open-Vocabulary Object Detection via Language Hierarchy
Open-Vocabulary Object Detection via Language Hierarchy Open
Recent studies on generalizable object detection have attracted increasing attention with additional weak supervision from large-scale datasets with image-level labels. However, weakly-supervised detection learning often suffers from image…
View article: Historical Test-time Prompt Tuning for Vision Foundation Models
Historical Test-time Prompt Tuning for Vision Foundation Models Open
Test-time prompt tuning, which learns prompts online with unlabelled test samples during the inference stage, has demonstrated great potential by learning effective prompts on-the-fly without requiring any task-specific annotations. Howeve…
View article: Accurate segmentation of intracellular organelle networks using low-level features and topological self-similarity
Accurate segmentation of intracellular organelle networks using low-level features and topological self-similarity Open
Motivation Intracellular organelle networks (IONs) such as the endoplasmic reticulum (ER) network and the mitochondrial (MITO) network serve crucial physiological functions. The morphology of these networks plays a critical role in mediati…
View article: Transcription factor EGR2 alleviates autoimmune uveitis via activation of GDF15 to modulate the retinal microglial phenotype
Transcription factor EGR2 alleviates autoimmune uveitis via activation of GDF15 to modulate the retinal microglial phenotype Open
Uveitis is a vision-threatening disease primarily driven by a dysregulated immune response, with retinal microglia playing a pivotal role in its progression. Although the transcription factor EGR2 is known to be closely associated with uve…
View article: Research Progress of Fluorescent Probes for Detection of Glutathione (GSH): Fluorophore, Photophysical Properties, Biological Applications
Research Progress of Fluorescent Probes for Detection of Glutathione (GSH): Fluorophore, Photophysical Properties, Biological Applications Open
Intracellular biothiols, including cysteine (Cys), glutathione (GSH), and homocysteine (Hcy), play a critical role in many physiological and pathological processes. Among them, GSH is the most abundant non-protein mercaptan (1–10 mM) in ce…
View article: A Survey on Evaluation of Multimodal Large Language Models
A Survey on Evaluation of Multimodal Large Language Models Open
Multimodal Large Language Models (MLLMs) mimic human perception and reasoning system by integrating powerful Large Language Models (LLMs) with various modality encoders (e.g., vision, audio), positioning LLMs as the "brain" and various mod…
View article: Deep learning and remote photoplethysmography powered advancements in contactless physiological measurement
Deep learning and remote photoplethysmography powered advancements in contactless physiological measurement Open
In recent decades, there has been ongoing development in the application of computer vision (CV) in the medical field. As conventional contact-based physiological measurement techniques often restrict a patient’s mobility in the clinical e…
View article: Celastrol exerts antiarrhythmic effects in chronic heart failure via NLRP3/Caspase-1/IL-1β signaling pathway
Celastrol exerts antiarrhythmic effects in chronic heart failure via NLRP3/Caspase-1/IL-1β signaling pathway Open
Celastrol could alleviate the adverse effects of VAs after MI partially by promoting autonomic nerve remodeling, ventricular electrical reconstruction and ion channel remodeling, and alleviating ventricular fibrosis and inflammatory respon…
View article: A feedback loop driven by H3K9 lactylation and HDAC2 in endothelial cells regulates VEGF-induced angiogenesis
A feedback loop driven by H3K9 lactylation and HDAC2 in endothelial cells regulates VEGF-induced angiogenesis Open
Background Vascular endothelial growth factor (VEGF) is one of the most powerful proangiogenic factors and plays an important role in multiple diseases. Increased glycolytic rates and lactate accumulation are associated with pathological a…
View article: Blockade of mesenteric and omental adipose tissue sensory neurons improves cardiac remodeling through sympathetic pathway
Blockade of mesenteric and omental adipose tissue sensory neurons improves cardiac remodeling through sympathetic pathway Open
Mesenteric and omental adipose tissue (MOAT) communicates directly with the heart through the secretion of bioactive molecules and indirectly through afferent signaling to the central nervous system. Myocardial infarction (MI) may induce p…
View article: Unveiling Trends: Nanoscale Materials Shaping Emerging Biomedical Applications
Unveiling Trends: Nanoscale Materials Shaping Emerging Biomedical Applications Open
The realm of biomedical materials continues to evolve rapidly, driven by innovative research across interdisciplinary domains. Leveraging big data from the CAS Content Collection, this study employs quantitative analysis through natural la…