Publications (27)
An alternative text representation to TF-IDF and Bag-of-Words
Zhixiang, Xu, Minmin Chen +2
In text mining, information retrieval, and machine learning, text documents are commonly represented through variants of sparse Bag of Words (sBoW) vectors (e.g. TF-IDF). Although…
Integrating Multi-scale and Multi-filtration Topological Features for Medical Image Classification
Pengfei Gu, Huimin Li, Haoteng Tang +6
Modern deep neural networks have shown remarkable performance in medical image classification. However, such networks either emphasize pixel-intensity features instead of fundament…
Cold freeze out of superheavy dark matter and Hubble tension
Zhijie, Xu
We propose a unified dark matter framework, the "X miracle", in which dark matter consists of superheavy, nonthermal X particles whose relic abundance is set by annihilation or dec…
On a critical acceleration scale of dark matter in Lambda-CDM and dynamical dark energy
Zhijie, Xu
Universal acceleration emerges in various empirical laws, yet its fundamental nature remains unclear. Using Illustris and Virgo N-body simulations, we propose is the sc…
Inverse energy cascade in self-gravitating collisionless dark matter flow and effects of halo shape
Zhijie, Xu
Halo-mediated mass and energy cascades are key to understand dark matter flow. Both cascades origin from the mass exchange between halo and out-of-halo sub-systems. Kinetic energy…
Inverse mass cascade in dark matter flow and effects on halo deformation, energy, size, and density profiles
Zhijie, Xu
Inverse mass cascade is a key feature of the intermediate statistically steady state for self-gravitating collisionless dark matter flow (SG-CFD). This paper focus on effects of ma…
Dark matter halo mass functions and density profiles from mass and energy cascade
Zhijie, Xu
Without relying on a spherical or ellipsoidal collapse model, we analytically derive the halo mass function and cuspy halo density (inner slope of -4/3) based on the mass and energ…
Whose Journey Matters? Investigating Identity Biases in Large Language Models (LLMs) for Travel Planning Assistance
Ruiping Ren, Yingwei, Xu +3
As large language models (LLMs) become increasingly integral to the hospitality and tourism industry, concerns about their fairness in serving diverse identity groups persist. Grou…
When Verification Hurts: Asymmetric Effects of Multi-Agent Feedback in Logic Proof Tutoring
Tahreem Yasir, Sutapa Dey Tithi, Benyamin Tabarsi +7
Large language models (LLMs) are increasingly used for automated tutoring, but their reliability in structured symbolic domains remains unclear. We study step-level feedback for pr…
Gemini 2.5: Pushing the Frontier with Advanced Reasoning, Multimodality, Long Context, and Next Generation Agentic Capabilities
Gheorghe Comanici, Eric Bieber, Mike Schaekermann +3431
In this report, we introduce the Gemini 2.X model family: Gemini 2.5 Pro and Gemini 2.5 Flash, as well as our earlier Gemini 2.0 Flash and Flash-Lite models. Gemini 2.5 Pro is our…
Can Deep Learning Assist Automatic Identification of Layered Pigments From XRF Data?
Bingjie, Xu, Yunan Wu +12
X-ray fluorescence spectroscopy (XRF) plays an important role for elemental analysis in a wide range of scientific fields, especially in cultural heritage. XRF imaging, which uses…
GMC Collisions As Triggers of Star Formation. IX. Chemical Evolution
Chia-Jung Hsu, Jonathan C. Tan, Jonathan Holdship +5
Collisions between giant molecular clouds (GMCs) have been proposed as a mechanism to trigger massive star and star cluster formation. To investigate the astrochemical signatures o…
Sliding-Window QPS (SW-QPS): A Perfect Parallel Iterative Switching Algorithm for Input-Queued Switches
Jingfan Meng, Long Gong, Jun +1
In this work, we first propose a parallel batch switching algorithm called Small-Batch Queue-Proportional Sampling (SB-QPS). Compared to other batch switching algorithms, SB-QPS si…
FreeRide: Harvesting Bubbles in Pipeline Parallelism
Jiashu Zhang, Zihan Pan, Molly +3
The occurrence of bubbles in pipeline parallelism is an inherent limitation that can account for more than 40% of the large language model (LLM) training time and is one of the mai…
Automated Discovery of Pairwise Interactions from Unstructured Data
Zuheng, Xu, Moksh Jain +5
Pairwise interactions between perturbations to a system can provide evidence for the causal dependencies of the underlying underlying mechanisms of a system. When observations are…
When Copilot Becomes Autopilot: Generative AI's Critical Risk to Knowledge Work and a Critical Solution
Advait Sarkar, Xiaotong, Xu +3
Generative AI, with its tendency to "hallucinate" incorrect results, may pose a risk to knowledge work by introducing errors. On the other hand, it may also provide unprecedented o…
Staircase Streaming for Low-Latency Multi-Agent Inference
Junlin Wang, Jue Wang, Zhen +5
Recent advances in large language models (LLMs) opened up new directions for leveraging the collective expertise of multiple LLMs. These methods, such as Mixture-of-Agents, typical…
Electric Vehicle Integration using Large-Scale Combined Transmission and Distribution Grid Models
Diana Wallison, Lyric Haylow, Jessica Wert +4
In this paper, we propose a unifying co-simulation framework integrating transportation demand, grid assets, land use, demographics, and emissions to optimally accelerate electric…
Electrically Controllable van der Waals Antiferromagnetic Spin Valve
Xue-Chao Zhai, Ziming, Xu +4
We propose a spin valve that is based on van der Waals antiferromagnetism and fully electrically controlled. The device is composed of two antiferromagnetic terminals that allow fo…
Towards impactful challenges: post-challenge paper, benchmarks and other dissemination actions
Antoine Marot, David Rousseau, Zhen +1
The conclusion of an AI challenge is not the end of its lifecycle; ensuring a long-lasting impact requires meticulous post-challenge activities. The long-lasting impact also needs…
High Current Density and Low Thermal Conductivity of Atomically Thin Semimetallic WTe2
Michal J. Mleczko, Runjie, Xu +6
Two-dimensional (2D) semimetals beyond graphene have been relatively unexplored in the atomically-thin limit. Here we introduce a facile growth mechanism for semimetallic WTe2 crys…
R(QPS-Serena) and R(QPS-Serenade): Two Novel Augmenting-Path Based Algorithms for Computing Approximate Maximum Weight Matching
Long Gong, Jun, Xu
In this addendum, we show that the switching algorithm QPS-SERENA can be converted R(QPS-SERENA), an algorithm for computing approximate Maximum Weight Matching (MWM). Empirically,…
Beyond Conservative Automated Driving in Multi-Agent Scenarios via Coupled Model Predictive Control and Deep Reinforcement Learning
Saeed Rahmani, Gözde Körpe, Zhenlin +4
Automated driving at unsignalized intersections is challenging due to complex multi-vehicle interactions and the need to balance safety and efficiency. Model Predictive Control (MP…
A Multimodal Adaptive Graph-based Intelligent Classification Model for Fake News
Jun-hao, Xu
Numerous studies have been proposed to detect fake news focusing on multi-modalities based on machine and/or deep learning. However, studies focusing on graph-based structures usin…
Energy Dissipation in Monolayer MoS Electronics
Eilam Yalon, Connor J. McClellan, Kirby K. H. Smithe +9
The advancement of nanoscale electronics has been limited by energy dissipation challenges for over a decade. Such limitations could be particularly severe for two-dimensional (2D)…
"It makes you think": Provocations Help Restore Critical Thinking to AI-Assisted Knowledge Work
Ian Drosos, Advait Sarkar, Xiaotong +2
Recent research suggests that the use of Generative AI tools may result in diminished critical thinking during knowledge work. We study the effect on knowledge work of provocations…
Bone tumor suppression in rabbits by hyperthermia below the clinical safety limit using aligned magnetic bone cement
Xiang Yu, Shan Gao, Dian Wu +12
Demonstrating highly efficient alternating current (AC) magnetic field heating of nanoparticles in physiological environments under clinically safe field parameters has remained a…