Bohan Li
YOU?
Author Swipe
View article: One View, Many Worlds: Single-Image to 3D Object Meets Generative Domain Randomization for One-Shot 6D Pose Estimation
One View, Many Worlds: Single-Image to 3D Object Meets Generative Domain Randomization for One-Shot 6D Pose Estimation Open
Estimating the 6D pose of arbitrary unseen objects from a single reference image is critical for robotics operating in the long-tail of real-world instances. However, this setting is notoriously challenging: 3D models are rarely available,…
View article: Three-dimensional spatial transcriptomics at isotropic resolution enabled by generative deep learning
Three-dimensional spatial transcriptomics at isotropic resolution enabled by generative deep learning Open
Mapping the complete three-dimensional (3D), transcriptome-wide spatial architecture of tissues and organs remains a fundamental challenge in biology. Typically, studies approximate 3D structures by profiling serial two-dimensional (2D) ti…
View article: TiDGRec: Dual-Graph Modeling with Target-intention Filtering for Session-based Recommendation
TiDGRec: Dual-Graph Modeling with Target-intention Filtering for Session-based Recommendation Open
Session-based recommendation (SBR) focuses on forecasting the next item a user is likely to select using brief and anonymous sequences of interactions. Existing methods face three key challenges: (1) difficulty in distinguishing noisy tran…
View article: MultiRAG: A Knowledge-Guided Framework for Mitigating Hallucination in Multi-Source Retrieval Augmented Generation
MultiRAG: A Knowledge-Guided Framework for Mitigating Hallucination in Multi-Source Retrieval Augmented Generation Open
Retrieval Augmented Generation (RAG) has emerged as a promising solution to address hallucination issues in Large Language Models (LLMs). However, the integration of multiple retrieval sources, while potentially more informative, introduce…
View article: Advancing personalized, predictive, and preventive medicine in bladder cancer: a multi-omics and machine learning approach for novel prognostic modeling, immune profiling, and therapeutic target discovery
Advancing personalized, predictive, and preventive medicine in bladder cancer: a multi-omics and machine learning approach for novel prognostic modeling, immune profiling, and therapeutic target discovery Open
Objective This study aimed to identify and analyze immunogenic cell death (ICD)-related multi-omics features in bladder cancer (BLCA) using single-cell RNA sequencing (scRNA-seq) and bulk RNA-seq data. By integrating these datasets, we sou…
View article: A harmonic current detection algorithm for aviation active power filter based on generalized delayed signal superposition
A harmonic current detection algorithm for aviation active power filter based on generalized delayed signal superposition Open
To address the limitations of traditional harmonic detection methods for active power filters in variable-frequency-grids of the More Electric Aircraft (MEA), including inadequate filtering performance and poor adaptability to frequency va…
View article: DiST-4D: Disentangled Spatiotemporal Diffusion with Metric Depth for 4D Driving Scene Generation
DiST-4D: Disentangled Spatiotemporal Diffusion with Metric Depth for 4D Driving Scene Generation Open
Current generative models struggle to synthesize dynamic 4D driving scenes that simultaneously support temporal extrapolation and spatial novel view synthesis (NVS) without per-scene optimization. A key challenge lies in finding an efficie…
View article: Enhancing Lightning Resilience: Predictive Models and Infrastructure Protection for UK Electric Power Systems
Enhancing Lightning Resilience: Predictive Models and Infrastructure Protection for UK Electric Power Systems Open
The UK’s goal of achieving net zero emissions by 2050 requires the construction of extensive new power infrastructure to accommodate low carbon energy technologies (e.g., offshore wind, nuclear) while mitigating climate risks. Lightn…
View article: MuDG: Taming Multi-modal Diffusion with Gaussian Splatting for Urban Scene Reconstruction
MuDG: Taming Multi-modal Diffusion with Gaussian Splatting for Urban Scene Reconstruction Open
Recent breakthroughs in radiance fields have significantly advanced 3D scene reconstruction and novel view synthesis (NVS) in autonomous driving. Nevertheless, critical limitations persist: reconstruction-based methods exhibit substantial …
View article: ReF Decompile: Relabeling and Function Call Enhanced Decompile
ReF Decompile: Relabeling and Function Call Enhanced Decompile Open
The goal of decompilation is to convert compiled low-level code (e.g., assembly code) back into high-level programming languages, enabling analysis in scenarios where source code is unavailable. This task supports various reverse engineeri…
View article: Special economic zones and inter-prefecture collaborative innovation: evidence from China
Special economic zones and inter-prefecture collaborative innovation: evidence from China Open
This paper studies whether and how the establishment of new special economic zones (SEZs) following a regulatory policy on SEZ construction can affect inter-prefecture collaborative innovation in China. This policy required SEZs to develop…
View article: The Use of Nonverbal Signs in Live Rock Band Performances
The Use of Nonverbal Signs in Live Rock Band Performances Open
Rock and Roll is a musical genre that emerged in the late 1940s United States. The music later developed into the popular genre of music that we are familiar with: Rock music. The music was almost always performed by a small group of peopl…
View article: LangSurf: Language-Embedded Surface Gaussians for 3D Scene Understanding
LangSurf: Language-Embedded Surface Gaussians for 3D Scene Understanding Open
Applying Gaussian Splatting to perception tasks for 3D scene understanding is becoming increasingly popular. Most existing works primarily focus on rendering 2D feature maps from novel viewpoints, which leads to an imprecise 3D language fi…
View article: MAGE: Multimodal Alignment and Generation Enhancement via Bridging Visual and Semantic Spaces
MAGE: Multimodal Alignment and Generation Enhancement via Bridging Visual and Semantic Spaces Open
In the latest advancements in multimodal learning, effectively addressing the spatial and semantic losses of visual data after encoding remains a critical challenge. This is because the performance of large multimodal models is positively …
View article: A Local Search Algorithm for MaxSMT(LIA)
A Local Search Algorithm for MaxSMT(LIA) Open
MaxSAT modulo theories (MaxSMT) is an important generalization of Satisfiability modulo theories (SMT) with various applications. In this paper, we focus on MaxSMT with the background theory of Linear Integer Arithmetic, denoted as MaxSMT(…
View article: ECEQ: Efficient Multi-source Contact Event Query Processing for Moving Objects
ECEQ: Efficient Multi-source Contact Event Query Processing for Moving Objects Open
Using trajectory data to query and analyze contact events is an emerging method for disease prevention and control. Existing contact event query processings only focus on single-source contact events (one-to-one), overlooking the more real…
View article: Joint beamforming and compressed sensing for uplink grant-free access
Joint beamforming and compressed sensing for uplink grant-free access Open
Compressed sensing (CS)-based techniques have been widely applied in the grant-free non-orthogonal multiple access (NOMA) to a single-antenna base station (BS). In this paper, we consider the multi-antenna reception at the BS for uplink gr…
View article: Closed-Loop Unsupervised Representation Disentanglement with $β$-VAE Distillation and Diffusion Probabilistic Feedback
Closed-Loop Unsupervised Representation Disentanglement with $β$-VAE Distillation and Diffusion Probabilistic Feedback Open
Representation disentanglement may help AI fundamentally understand the real world and thus benefit both discrimination and generation tasks. It currently has at least three unresolved core issues: (i) heavy reliance on label annotation an…
View article: Numerical Simulation Study on Flow Field Characteristics and Separation Efficiency of Micro Cyclone by Shunt Ratio
Numerical Simulation Study on Flow Field Characteristics and Separation Efficiency of Micro Cyclone by Shunt Ratio Open
In this study, based on the current situation of waste drilling fluid treatment, the operating parameters of micro cyclones are investigated. A 10-mm micro cyclone used to separate ultrafine particles was designed and the effect of the spl…
View article: Comprehensive Resource Treatment of Titanium White Waste Acid by Chlorination Method Based on Mechanical Separation Method
Comprehensive Resource Treatment of Titanium White Waste Acid by Chlorination Method Based on Mechanical Separation Method Open
Using the chlorination method to produce titanium dioxide, approximately 6~8t of waste acid with a mass fraction of about 20% is generated per 1t of titanium dioxide produced.In order to address the issue of acid waste pollution and resour…
View article: CFD-based study of inlet flow rate variation on flow field of micro-cyclone
CFD-based study of inlet flow rate variation on flow field of micro-cyclone Open
In this study, the operational parameters of the micro-cyclone were investigated. A 10 mm micro-cyclone suitable for waste drilling mud treatment was designed and the effect of the inlet on the flow field within the micro-cyclone as well a…
View article: ODIN: Object Density Aware Index for CkNN Queries over Moving Objects on Road Networks
ODIN: Object Density Aware Index for CkNN Queries over Moving Objects on Road Networks Open
We study the problem of processing continuous k nearest neighbor (CkNN) queries over moving objects on road networks, which is an essential operation in a variety of applications. We are particularly concerned with scenarios where the obje…
View article: Self-Supervised Dynamic Hypergraph Recommendation based on Hyper-Relational Knowledge Graph
Self-Supervised Dynamic Hypergraph Recommendation based on Hyper-Relational Knowledge Graph Open
Knowledge graphs (KGs) are commonly used as side information to enhance collaborative signals and improve recommendation quality. In the context of knowledge-aware recommendation (KGR), graph neural networks (GNNs) have emerged as promisin…
View article: A One-Size-Fits-Three Representation Learning Framework for Patient Similarity Search
A One-Size-Fits-Three Representation Learning Framework for Patient Similarity Search Open
Patient similarity search is an essential task in healthcare. Recent studies adopted electronic health records (EHRs) to learn patient representations for measuring the clinical similarities. These methods outperformed traditional methods,…
View article: Detection of Reflected XSS Vulnerabilities Based on Paths-Attention Method
Detection of Reflected XSS Vulnerabilities Based on Paths-Attention Method Open
Cross-site scripting vulnerability (XSS) is one of the most frequently exploited and harmful vulnerabilities among web vulnerabilities. In recent years, many researchers have used different machine learning methods to detect network attack…
View article: A Survey of Advanced Information Fusion System: from Model-Driven to Knowledge-Enabled
A Survey of Advanced Information Fusion System: from Model-Driven to Knowledge-Enabled Open
Advanced knowledge engineering (KE), represented by knowledge graph (KG), drives the development of various fields and engineering technologies and provides various knowledge fusion and knowledge empowerment interfaces. At the same time, a…