Truyen Tran
YOU?
Author Swipe
View article: Rethinking Deep Alignment Through The Lens Of Incomplete Learning
Rethinking Deep Alignment Through The Lens Of Incomplete Learning Open
Large language models exhibit systematic vulnerabilities to adversarial attacks despite extensive safety alignment. We provide a mechanistic analysis revealing that position-dependent gradient weakening during autoregressive training creat…
View article: Planner-Refiner: Dynamic Space-Time Refinement for Vision-Language Alignment in Videos
Planner-Refiner: Dynamic Space-Time Refinement for Vision-Language Alignment in Videos Open
Vision-language alignment in video must address the complexity of language, evolving interacting entities, their action chains, and semantic gaps between language and vision. This work introduces Planner-Refiner, a framework to overcome th…
View article: On the metallogeny of gold and its associated metals of the Truong Son orogenic belt and adjacent territories, Vietnam
On the metallogeny of gold and its associated metals of the Truong Son orogenic belt and adjacent territories, Vietnam Open
It is demonstrated that the Truong Son orogenic belt possesses a substantially Au-Cu metallogenic profile with a subordinate role of Sn-Li mineralization. The mineralogenetic stages of different ages are estimated in significance. The impo…
View article: Impact of chirality on nanotube fracture strain: comprehensive machine learning potential calculations
Impact of chirality on nanotube fracture strain: comprehensive machine learning potential calculations Open
We present a comprehensive study of fracture strains in carbon nanotubes (CNTs) as a function of their chirality using the direct integration of the exchange potential (DIEP) machine learning potential model. By simulating fracture using q…
View article: Physics-informed Machine Learning for Medical Image Analysis
Physics-informed Machine Learning for Medical Image Analysis Open
The incorporation of physical information in machine learning frameworks is transforming medical image analysis (MIA). Integrating fundamental knowledge and governing physical laws not only improves analysis performance but also enhances t…
View article: Now or Never: Exploring Factors Influencing Online Impulsive Buying Behavior on Fashion Brand Websites among Generation Z Consumers
Now or Never: Exploring Factors Influencing Online Impulsive Buying Behavior on Fashion Brand Websites among Generation Z Consumers Open
Recently, online shopping has become increasingly popular, and one of the most favored product categories is fashion. This study investigates the factors influencing impulsive purchasing on fashion brand websites among Generation Z in Hano…
View article: Planner-Refiner: Dynamic Space-Time Refinement for Vision-Language Alignment in Videos
Planner-Refiner: Dynamic Space-Time Refinement for Vision-Language Alignment in Videos Open
Vision-language alignment in video must address the complexity of language, evolving interacting entities, their action chains, and semantic gaps between language and vision. This work introduces Planner-Refiner, a framework to overcome th…
View article: Towards universal early screening for cerebral palsy: a roadmap for automated General Movements Assessment
Towards universal early screening for cerebral palsy: a roadmap for automated General Movements Assessment Open
Cerebral Palsy Foundation, Cerebral Palsy Alliance, European Union Born to Get There, the National Health and Medical Research Council.
View article: Improving Disordered Crystal Modeling with Nonlinear Atomic Embedding for Superconducting Materials
Improving Disordered Crystal Modeling with Nonlinear Atomic Embedding for Superconducting Materials Open
Accurately predicting properties like the superconducting critical temperature (T$_c$) is crucial for materials discovery, yet it remains challenging for materials exhibiting crystal disorder. Conventional deep learning models often repres…
View article: Bộ Chuyển Đổi DC-DC Tăng Áp Mới Dựa Trên Cấu Hình Chuyển Tụ với Độ Lợi Điện Áp Cao
Bộ Chuyển Đổi DC-DC Tăng Áp Mới Dựa Trên Cấu Hình Chuyển Tụ với Độ Lợi Điện Áp Cao Open
Bài báo này trình bày một cấu hình DC-DC tăng áp có khả năng cung cấp một điện áp đầu ra DC cao dựa trên kỹ thuật chuyển tụ SC (Switch Capacitor - SC), được gọi là cấu hình tăng áp dựa trên kỹ thuật chuyển tụ (SCBC). Cấu hình SCBC được xây…
View article: Motivational Teaching Strategies:
Motivational Teaching Strategies: Open
This study investigates EFL teachers’ and students’ perceptions of motivational teaching strategies at a vocational school in Khanh Hoa Province. Using a mixed-methods approach, data were collected from five teachers and 300 students throu…
View article: Learning Structural Causal Models from Ordering: Identifiable Flow Models
Learning Structural Causal Models from Ordering: Identifiable Flow Models Open
In this study, we address causal inference when only observational data and a valid causal ordering from the causal graph are available. We introduce a set of flow models that can recover component-wise, invertible transformation of exogen…
View article: Progressive Multi-granular Alignments for Grounded Reasoning in Large Vision-Language Models
Progressive Multi-granular Alignments for Grounded Reasoning in Large Vision-Language Models Open
Existing Large Vision-Language Models (LVLMs) excel at matching concepts across multi-modal inputs but struggle with compositional concepts and high-level relationships between entities. This paper introduces Progressive multi-granular Vis…
View article: LaGR-SEQ: Language-guided reinforcement learning with sample-efficient querying
LaGR-SEQ: Language-guided reinforcement learning with sample-efficient querying Open
Large language models (LLMs) have recently demonstrated their impressive ability to provide context-aware responses via text. This ability could potentially be used to predict plausible solutions in sequential decision making tasks pertain…
View article: S2MoE: Robust Sparse Mixture of Experts via Stochastic Learning
S2MoE: Robust Sparse Mixture of Experts via Stochastic Learning Open
Sparse Mixture of Experts (SMoE) enables efficient training of large language models by routing input tokens to a select number of experts. However, training SMoE remains challenging due to the issue of representation collapse. Recent stud…
View article: PID3Net: a deep learning approach for single-shot coherent X-ray diffraction imaging of dynamic phenomena
PID3Net: a deep learning approach for single-shot coherent X-ray diffraction imaging of dynamic phenomena Open
This paper introduces a deep learning (DL)-based method for phase retrieval tailored to single-shot, multiple-frame coherent X-ray diffraction imaging (CXDI), designed specifically for visualizing local nanostructural dynamics within a lar…
View article: Finding the Trigger: Causal Abductive Reasoning on Video Events
Finding the Trigger: Causal Abductive Reasoning on Video Events Open
This paper introduces a new problem, Causal Abductive Reasoning on Video Events (CARVE), which involves identifying causal relationships between events in a video and generating hypotheses about causal chains that account for the occurrenc…
View article: Scale matters: simulation of nanoscopic dendrite initiation in the lithium solid electrolyte interphase using a machine learning potential
Scale matters: simulation of nanoscopic dendrite initiation in the lithium solid electrolyte interphase using a machine learning potential Open
Although lithium solid state electrolytes promise to mitigate the chemical instabilities of liquid electrolytes in today's mainstream rechargeable batteries, solid state electrolytes still suffer from dendrite formation which leads to batt…
View article: Efficient Symmetry-Aware Materials Generation via Hierarchical Generative Flow Networks
Efficient Symmetry-Aware Materials Generation via Hierarchical Generative Flow Networks Open
Discovering new solid-state materials requires rapidly exploring the vast space of crystal structures and locating stable regions. Generating stable materials with desired properties and compositions is extremely difficult as we...
View article: SimSMoE: Toward Efficient Training Mixture of Experts via Solving Representational Collapse
SimSMoE: Toward Efficient Training Mixture of Experts via Solving Representational Collapse Open
View article: Understanding the Variables Driving India’s Tourism: An Exploratory Study
Understanding the Variables Driving India’s Tourism: An Exploratory Study Open
Tourism significantly contributes to India’s economic growth, employment generation, and cultural exchange. This study explores the key factors influencing the development of India’s tourism industry, using secondary data and qualitative a…
View article: Learning Structural Causal Models from Ordering: Identifiable Flow Models
Learning Structural Causal Models from Ordering: Identifiable Flow Models Open
In this study, we address causal inference when only observational data and a valid causal ordering from the causal graph are available. We introduce a set of flow models that can recover component-wise, invertible transformation of exogen…
View article: Progressive Multi-granular Alignments for Grounded Reasoning in Large Vision-Language Models
Progressive Multi-granular Alignments for Grounded Reasoning in Large Vision-Language Models Open
Existing Large Vision-Language Models (LVLMs) excel at matching concepts across multi-modal inputs but struggle with compositional concepts and high-level relationships between entities. This paper introduces Progressive multi-granular Vis…
View article: MP-PINN: A Multi-Phase Physics-Informed Neural Network for Epidemic Forecasting
MP-PINN: A Multi-Phase Physics-Informed Neural Network for Epidemic Forecasting Open
Forecasting temporal processes such as virus spreading in epidemics often requires more than just observed time-series data, especially at the beginning of a wave when data is limited. Traditional methods employ mechanistic models like the…
View article: Efficient Symmetry-Aware Materials Generation via Hierarchical Generative Flow Networks
Efficient Symmetry-Aware Materials Generation via Hierarchical Generative Flow Networks Open
Discovering new solid-state materials requires rapidly exploring the vast space of crystal structures and locating stable regions. Generating stable materials with desired properties and compositions is extremely difficult as we search for…
View article: Scale matters: simulation of nanoscopic dendrite initiation in the lithium solid electrolyte interphase using a machine learning potential
Scale matters: simulation of nanoscopic dendrite initiation in the lithium solid electrolyte interphase using a machine learning potential Open
Although lithium solid state electrolytes promise to mitigate the chemical instabilities of liquid electrolytes in today's mainstream rechargeable batteries, solid state electrolytes still suffer from dendrite formation which leads to batt…
View article: Correction: Predicting the risk of diabetes complications using machine learning and social administrative data in a country with ethnic inequities in health: Aotearoa New Zealand
Correction: Predicting the risk of diabetes complications using machine learning and social administrative data in a country with ethnic inequities in health: Aotearoa New Zealand Open
View article: Predicting the risk of diabetes complications using machine learning and social administrative data in a country with ethnic inequities in health: Aotearoa New Zealand
Predicting the risk of diabetes complications using machine learning and social administrative data in a country with ethnic inequities in health: Aotearoa New Zealand Open
We provide further evidence that ML with administrative health data can be used for meaningful future prediction of health outcomes. As such, it could be utilized to inform health planning and healthcare resource allocation for diabetes ma…
View article: The impact of Firm Generated Content (FGC) and User Generated Content (UGC) on the intention to use telehealth services
The impact of Firm Generated Content (FGC) and User Generated Content (UGC) on the intention to use telehealth services Open
The purpose of this paper is to explore the ability of the combination of Theory of Reasoned Action (TRA) and Information Adoption Model (IAM); to predict and explain how Firm-Generated Content (FGC) and User-Generated Content (UGC) influe…
View article: Unified Framework with Consistency across Modalities for Human Activity Recognition
Unified Framework with Consistency across Modalities for Human Activity Recognition Open
Recognizing human activities in videos is challenging due to the spatio-temporal complexity and context-dependence of human interactions. Prior studies often rely on single input modalities, such as RGB or skeletal data, limiting their abi…