Bingsheng He
YOU?
Author Swipe
View article: PaperDebugger: A Plugin-Based Multi-Agent System for In-Editor Academic Writing, Review, and Editing
PaperDebugger: A Plugin-Based Multi-Agent System for In-Editor Academic Writing, Review, and Editing Open
Large language models are increasingly embedded into academic writing workflows, yet existing assistants remain external to the editor, preventing deep interaction with document state, structure, and revision history. This separation makes…
View article: Blurred Encoding for Trajectory Representation Learning
Blurred Encoding for Trajectory Representation Learning Open
Trajectory representation learning (TRL) maps trajectories to vector embeddings and facilitates tasks such as trajectory classification and similarity search. State-of-the-art (SOTA) TRL methods transform raw GPS trajectories to grid or ro…
View article: A comprehensive taxonomy of prompt engineering techniques for large language models
A comprehensive taxonomy of prompt engineering techniques for large language models Open
Large Language Models (LLMs) have demonstrated remarkable performance across various downstream tasks, as evidenced by numerous studies. Since 2022, generative AI has shown significant potential in diverse application domains, including ga…
View article: Towards Unsupervised Open-Set Graph Domain Adaptation via Dual Reprogramming
Towards Unsupervised Open-Set Graph Domain Adaptation via Dual Reprogramming Open
Unsupervised Graph Domain Adaptation has become a promising paradigm for transferring knowledge from a fully labeled source graph to an unlabeled target graph. Existing graph domain adaptation models primarily focus on the closed-set setti…
View article: LLM DNA: Tracing Model Evolution via Functional Representations
LLM DNA: Tracing Model Evolution via Functional Representations Open
The explosive growth of large language models (LLMs) has created a vast but opaque landscape: millions of models exist, yet their evolutionary relationships through fine-tuning, distillation, or adaptation are often undocumented or unclear…
View article: Disagreements in Reasoning: How a Model's Thinking Process Dictates Persuasion in Multi-Agent Systems
Disagreements in Reasoning: How a Model's Thinking Process Dictates Persuasion in Multi-Agent Systems Open
The rapid proliferation of recent Multi-Agent Systems (MAS), where Large Language Models (LLMs) and Large Reasoning Models (LRMs) usually collaborate to solve complex problems, necessitates a deep understanding of the persuasion dynamics t…
View article: TempGraph: An Efficient Chain-driven Temporal Graph Computing Framework on the GPU
TempGraph: An Efficient Chain-driven Temporal Graph Computing Framework on the GPU Open
View article: Blurred Encoding for Trajectory Representation Learning
Blurred Encoding for Trajectory Representation Learning Open
View article: ML-Asset Management: Curation, Discovery, and Utilization
ML-Asset Management: Curation, Discovery, and Utilization Open
Machine learning (ML) assets, such as models, datasets, and metadata—are central to modern ML workflows. Despite their explosive growth in practice, these assets are often underutilized due to fragmented documentation, siloed storage, inco…
View article: Towards Evaluting Fake Reasoning Bias in Language Models
Towards Evaluting Fake Reasoning Bias in Language Models Open
Large Reasoning Models (LRMs), evolved from standard Large Language Models (LLMs), are increasingly utilized as automated judges because of their explicit reasoning processes. Yet we show that both LRMs and standard LLMs are vulnerable to …
View article: NTSFormer: A Self-Teaching Graph Transformer for Multimodal Isolated Cold-Start Node Classification
NTSFormer: A Self-Teaching Graph Transformer for Multimodal Isolated Cold-Start Node Classification Open
Isolated cold-start node classification on multimodal graphs is challenging because such nodes have no edges and often have missing modalities (e.g., absent text or image features). Existing methods address structural isolation by degradin…
View article: HLStrans: Dataset for C-to-HLS Hardware Code Synthesis
HLStrans: Dataset for C-to-HLS Hardware Code Synthesis Open
High-Level Synthesis (HLS) enables hardware design from C/C++ kernels but requires extensive transformations, such as restructuring code, inserting pragmas, adapting data types, and repairing non-synthesizable constructs, to achieve effici…
View article: Dupin: A Parallel Framework for Densest Subgraph Discovery in Fraud Detection on Massive Graphs
Dupin: A Parallel Framework for Densest Subgraph Discovery in Fraud Detection on Massive Graphs Open
Detecting fraudulent activities in financial and e-commerce transaction networks is crucial. One effective method for this is Densest Subgraph Discovery (DSD). However, deploying DSD methods in production systems faces substantial scalabil…
View article: Community Detection in Heterogeneous Information Networks Without Materialization
Community Detection in Heterogeneous Information Networks Without Materialization Open
Community detection in heterogeneous information networks (HINs) poses significant challenges due to the diversity of entity types and the complexity of their interrelations. While traditional algorithms may perform adequately in some scen…
View article: Clementi: Efficient Load Balancing and Communication Overlap for Multi-FPGA Graph Processing
Clementi: Efficient Load Balancing and Communication Overlap for Multi-FPGA Graph Processing Open
Efficient graph processing is critical in various modern applications, such as social network analysis, recommendation systems, and large-scale data mining. Traditional single-FPGA systems struggle to handle the increasing size and complex…
View article: Stereotactic body radiation therapy (SBRT) increases anti-PD-1 antitumor activity by enhancing the tumor immune microenvironment in mice with metastatic hepatocellular carcinoma
Stereotactic body radiation therapy (SBRT) increases anti-PD-1 antitumor activity by enhancing the tumor immune microenvironment in mice with metastatic hepatocellular carcinoma Open
This manuscript was previously submitted as a preprint only in Experimental Hematology & Oncology and has no conflict of interest with this submission.
View article: Adaptive Inertial Method
Adaptive Inertial Method Open
In this paper, we introduce the Adaptive Inertial Method (AIM), a novel framework for accelerated first-order methods through a customizable inertial term. We provide a rigorous convergence analysis establishing a global convergence rate o…
View article: Option Pricing Based on Several Monte Carlo Techniques
Option Pricing Based on Several Monte Carlo Techniques Open
The Monte Carlo method is broadly used in financial technology and engineering for pricing complex derivatives and managing risk due to its flexibility and adaptability. However, Monte Carlo simulation may suffer from high variance problem…
View article: Aggregate to Adapt: Node-Centric Aggregation for Multi-Source-Free Graph Domain Adaptation
Aggregate to Adapt: Node-Centric Aggregation for Multi-Source-Free Graph Domain Adaptation Open
View article: Dupin: A Parallel Framework for Densest Subgraph Discovery in Fraud Detection on Massive Graphs (Technical Report)
Dupin: A Parallel Framework for Densest Subgraph Discovery in Fraud Detection on Massive Graphs (Technical Report) Open
Detecting fraudulent activities in financial and e-commerce transaction networks is crucial. One effective method for this is Densest Subgraph Discovery (DSD). However, deploying DSD methods in production systems faces substantial scalabil…
View article: Modality-Independent Graph Neural Networks with Global Transformers for Multimodal Recommendation
Modality-Independent Graph Neural Networks with Global Transformers for Multimodal Recommendation Open
Multimodal recommendation systems can learn users' preferences from existing user-item interactions as well as the semantics of multimodal data associated with items. Many existing methods model this through a multimodal user-item graph, a…
View article: N <scp>ode</scp> I <scp>mport</scp> : Imbalanced Node Classification with Node Importance Assessment
N <span>ode</span> I <span>mport</span> : Imbalanced Node Classification with Node Importance Assessment Open
View article: ScalaGBM: Memory Efficient GBDT Training for High-Dimensional Data on GPU
ScalaGBM: Memory Efficient GBDT Training for High-Dimensional Data on GPU Open
View article: FeatInsight: An Online ML Feature Management System on 4Paradigm Sage-Studio Platform
FeatInsight: An Online ML Feature Management System on 4Paradigm Sage-Studio Platform Open
Feature management is essential for many online machine learning applications and can often become the performance bottleneck (e.g., taking up to 70% of the overall latency in sales prediction service). Improper feature configurations (e.g…
View article: JudgeLRM: Large Reasoning Models as a Judge
JudgeLRM: Large Reasoning Models as a Judge Open
Large Language Models (LLMs) are increasingly adopted as evaluators, offering a scalable alternative to human annotation. However, existing supervised fine-tuning (SFT) approaches often fall short in domains that demand complex reasoning. …
View article: PyGDA: A Python Library for Graph Domain Adaptation
PyGDA: A Python Library for Graph Domain Adaptation Open
Graph domain adaptation has emerged as a promising approach to facilitate knowledge transfer across different domains. Recently, numerous models have been proposed to enhance their generalization capabilities in this field. However, there …
View article: FpgaHub: Fpga-centric Hyper-heterogeneous Computing Platform for Big Data Analytics
FpgaHub: Fpga-centric Hyper-heterogeneous Computing Platform for Big Data Analytics Open
Modern data analytics requires a huge amount of computing power and processes a massive amount of data. At the same time, the underlying computing platform is becoming much more heterogeneous on both hardware and software. Even though spec…
View article: Helios: Efficient Distributed Dynamic Graph Sampling for Online GNN Inference
Helios: Efficient Distributed Dynamic Graph Sampling for Online GNN Inference Open
View article: The Lottery LLM Hypothesis, Rethinking What Abilities Should LLM Compression Preserve?
The Lottery LLM Hypothesis, Rethinking What Abilities Should LLM Compression Preserve? Open
Motivated by reducing the computational and storage costs of LLMs, model compression and KV cache compression have attracted much attention from researchers. However, current methods predominantly emphasize maintaining the performance of c…
View article: Revisiting the Design of In-Memory Dynamic Graph Storage
Revisiting the Design of In-Memory Dynamic Graph Storage Open
The effectiveness of in-memory dynamic graph storage (DGS) for supporting concurrent graph read and write queries is crucial for real-time graph analytics and updates. Various methods have been proposed, for example, LLAMA, Aspen, LiveGrap…