Ming‐Wei Chang
YOU?
Author Swipe
View article: Bone marrow mesenchymal stem cell-loaded HAMA hydrogel within highly bionic nerve guidance conduits for peripheral nerve regeneration
Bone marrow mesenchymal stem cell-loaded HAMA hydrogel within highly bionic nerve guidance conduits for peripheral nerve regeneration Open
Nerve guidance conduits (NGCs) effectively support and guide the regeneration of injured nerves. However, traditional NGCs often lack essential growth factors and fail to create a biomimetic microenvironment conducive to nerve regrowth. Th…
View article: The Role of Perceived Benefits in Buffering Gastrointestinal-Symptom Burden Among Post-Operative Colorectal Cancer Patients: A Six-Month Longitudinal Study
The Role of Perceived Benefits in Buffering Gastrointestinal-Symptom Burden Among Post-Operative Colorectal Cancer Patients: A Six-Month Longitudinal Study Open
Objectives: The goal of this longitudinal study was to explore how gastrointestinal-symptom distress and benefit finding together influence health-related quality of life (HRQOL) over time in colorectal cancer (CRC) survivors. Colorectal c…
View article: Magnetic Field‐Assisted Conductive Nerve Guidance Conduit Enabling Peripheral Nerve Regeneration with Wireless Electrical Stimulation
Magnetic Field‐Assisted Conductive Nerve Guidance Conduit Enabling Peripheral Nerve Regeneration with Wireless Electrical Stimulation Open
Nerve guidance conduits capable of wireless stimulation represent a promising approach for addressing peripheral nerve defects. However, traditional electrical stimulation methods are not sufficiently convenient and may cause secondary dam…
View article: KITTEN: A Knowledge-Intensive Evaluation of Image Generation on Visual Entities
KITTEN: A Knowledge-Intensive Evaluation of Image Generation on Visual Entities Open
Recent advances in text-to-image generation have improved the quality of synthesized images, but evaluations mainly focus on aesthetics or alignment with text prompts. Thus, it remains unclear whether these models can accurately represent …
View article: Imagen 3
Imagen 3 Open
We introduce Imagen 3, a latent diffusion model that generates high quality images from text prompts. We describe our quality and responsibility evaluations. Imagen 3 is preferred over other state-of-the-art (SOTA) models at the time of ev…
View article: Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More?
Can Long-Context Language Models Subsume Retrieval, RAG, SQL, and More? Open
Long-context language models (LCLMs) have the potential to revolutionize our approach to tasks traditionally reliant on external tools like retrieval systems or databases. Leveraging LCLMs' ability to natively ingest and process entire cor…
View article: Gecko: Versatile Text Embeddings Distilled from Large Language Models
Gecko: Versatile Text Embeddings Distilled from Large Language Models Open
We present Gecko, a compact and versatile text embedding model. Gecko achieves strong retrieval performance by leveraging a key idea: distilling knowledge from large language models (LLMs) into a retriever. Our two-step distillation proces…
View article: MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions
MagicLens: Self-Supervised Image Retrieval with Open-Ended Instructions Open
Image retrieval, i.e., finding desired images given a reference image, inherently encompasses rich, multi-faceted search intents that are difficult to capture solely using image-based measures. Recent works leverage text instructions to al…
View article: Pioneering Astaxanthin-Tumor Cell Membrane Nanoparticles for Innovative Targeted Drug Delivery on Melanoma
Pioneering Astaxanthin-Tumor Cell Membrane Nanoparticles for Innovative Targeted Drug Delivery on Melanoma Open
Our findings illustrated the potential of astaxanthin nanovesicles as an anticancer drug.
View article: Leveraging Oriented Lateral Walls of Nerve Guidance Conduit with Core–Shell MWCNTs Fibers for Peripheral Nerve Regeneration
Leveraging Oriented Lateral Walls of Nerve Guidance Conduit with Core–Shell MWCNTs Fibers for Peripheral Nerve Regeneration Open
Peripheral nerve regeneration and functional recovery rely on the chemical, physical, and structural properties of nerve guidance conduits (NGCs). However, the limited support for long‐distance nerve regeneration and axonal guidance has hi…
View article: Instruct-Imagen: Image Generation with Multi-modal Instruction
Instruct-Imagen: Image Generation with Multi-modal Instruction Open
This paper presents instruct-imagen, a model that tackles heterogeneous image generation tasks and generalizes across unseen tasks. We introduce *multi-modal instruction* for image generation, a task representation articulating a range of …
View article: Fabrication of miconazole nitrate solid lipid nanoparticle loaded microneedle patches for the treatment of <i>Candida albicans</i> biofilms
Fabrication of miconazole nitrate solid lipid nanoparticle loaded microneedle patches for the treatment of <i>Candida albicans</i> biofilms Open
Delivery of miconazole nitrate into deep skin layers via SLN loaded MN patches resulted in the eradication of fungal biofilms and wound healing.
View article: Single-atomic site catalysts for electrochemical nitrogen fixation
Single-atomic site catalysts for electrochemical nitrogen fixation Open
Renewable-energy-derived electrochemical nitrogen fixation represents a sustainable way to produce green ammonia (NH3), but the energy efficiency is limited by its sluggish kinetics and complex reaction pathways. Highly active, selective a…
View article: QUEST: A Retrieval Dataset of Entity-Seeking Queries with Implicit Set Operations
QUEST: A Retrieval Dataset of Entity-Seeking Queries with Implicit Set Operations Open
Formulating selective information needs results in queries that implicitly specify set operations, such as intersection, union, and difference. For instance, one might search for "shorebirds that are not sandpipers" or "science-fiction fil…
View article: Graphene-Based Flexible Sensors for Respiratory and Airflow Monitoring
Graphene-Based Flexible Sensors for Respiratory and Airflow Monitoring Open
Flexible devices have been rapidly developed and applied in various applications. However, there have been few reports on printable graphene-based sensors with customized structures and properties capable of respiratory and airflow monitor…
View article: Conditional Adapters: Parameter-efficient Transfer Learning with Fast Inference
Conditional Adapters: Parameter-efficient Transfer Learning with Fast Inference Open
We propose Conditional Adapter (CoDA), a parameter-efficient transfer learning method that also improves inference efficiency. CoDA generalizes beyond standard adapter approaches to enable a new way of balancing speed and accuracy using co…
View article: Polycaprolactone/multi-walled carbon nanotube nerve guidance conduits with tunable channels fabricated via novel extrusion-stretching method for peripheral nerve repair
Polycaprolactone/multi-walled carbon nanotube nerve guidance conduits with tunable channels fabricated via novel extrusion-stretching method for peripheral nerve repair Open
Multi-channeled nerve guidance conduit is a prospective way to repair peripheral nerve injury, which is still difficult to be fabricated. A novel extrusion-stretching method was utilized in this study to produce multi-walled carbon nanotub…
View article: Rethinking the Role of Token Retrieval in Multi-Vector Retrieval
Rethinking the Role of Token Retrieval in Multi-Vector Retrieval Open
Multi-vector retrieval models such as ColBERT [Khattab and Zaharia, 2020] allow token-level interactions between queries and documents, and hence achieve state of the art on many information retrieval benchmarks. However, their non-linear …
View article: Subject-driven Text-to-Image Generation via Apprenticeship Learning
Subject-driven Text-to-Image Generation via Apprenticeship Learning Open
Recent text-to-image generation models like DreamBooth have made remarkable progress in generating highly customized images of a target subject, by fine-tuning an ``expert model'' for a given subject from a few examples. However, this proc…
View article: Can Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions?
Can Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions? Open
Pre-trained vision and language models have demonstrated state-of-the-art capabilities over existing tasks involving images and texts, including visual question answering. However, it remains unclear whether these models possess the capabi…
View article: Open-domain Visual Entity Recognition: Towards Recognizing Millions of Wikipedia Entities
Open-domain Visual Entity Recognition: Towards Recognizing Millions of Wikipedia Entities Open
Large-scale multi-modal pre-training models such as CLIP and PaLI exhibit strong generalization on various visual domains and tasks. However, existing image classification benchmarks often evaluate recognition on a specific domain (e.g., o…
View article: QUEST: A Retrieval Dataset of Entity-Seeking Queries with Implicit Set Operations
QUEST: A Retrieval Dataset of Entity-Seeking Queries with Implicit Set Operations Open
Formulating selective information needs results in queries that implicitly specify set operations, such as intersection, union, and difference. For instance, one might search for “shorebirds that are not sandpipers” or “science-fiction fil…
View article: Can Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions?
Can Pre-trained Vision and Language Models Answer Visual Information-Seeking Questions? Open
Pre-trained vision and language models have demonstrated state-of-the-art capabilities over existing tasks involving images and texts, including visual question answering. However, it remains unclear whether these models possess the capabi…
View article: Meta-Learning Fast Weight Language Models
Meta-Learning Fast Weight Language Models Open
Dynamic evaluation of language models (LMs) adapts model parameters at test time using gradient information from previous tokens and substantially improves LM performance. However, it requires over 3x more compute than standard inference. …