Publications (33)
Towards Scalable Backpropagation-Free Gradient Estimation
Daniel Wang, Evan Markou, Dylan Campbell
While backpropagation--reverse-mode automatic differentiation--has been extraordinarily successful in deep learning, it requires two passes (forward and backward) through the neura…
Isolated Wolf-Rayet Stars and O Supergiants in the Galactic Center Region Identified via Paschen-alpha Excess
Jon Mauerhan, Angela Cotera, Hui Dong +4
We report the discovery of 19 hot, evolved, massive stars near the Galactic center region (GCR). These objects were selected for spectroscopy owing to their detection as strong sou…
NeuroBind: Towards Unified Multimodal Representations for Neural Signals
Fengyu Yang, Chao Feng, Daniel Wang +8
Understanding neural activity and information representation is crucial for advancing knowledge of brain function and cognition. Neural activity, measured through techniques like e…
SWE-Marathon: Can Agents Autonomously Complete Ultra-Long-Horizon Software Work?
Rishi Desai, Jesse Hu, Joan Cabezas +23
AI agents are increasingly expected to complete long-horizon workflows that require sustained progress over hours, millions of tokens, and complex environments. Yet current agent b…
HOMER: Homography-Based Efficient Multi-view 3D Object Removal
Jingcheng Ni, Weiguang Zhao, Daniel Wang +4
3D object removal is an important sub-task in 3D scene editing, with broad applications in scene understanding, augmented reality, and robotics. However, existing methods struggle…
ODE-GS: Latent ODEs for Dynamic Scene Extrapolation with 3D Gaussian Splatting
Daniel Wang, Patrick Rim, Tian Tian +3
We introduce ODE-GS, a novel approach that integrates 3D Gaussian Splatting with latent neural ordinary differential equations (ODEs) to enable future extrapolation of dynamic 3D s…
X-ray spectroscopy of the hot has in the M31 bulge
Jiren Liu, Daniel Wang, Zhiyuan Li +1
We present an X-ray spectroscopic study of the nuclear region of the M31 bulge, based on observations of the Xmm-Newton Reflection Grating Spectrometers. The obtained high-resoluti…
RSA: Resolving Scale Ambiguities in Monocular Depth Estimators through Language Descriptions
Ziyao Zeng, Yangchao Wu, Hyoungseob Park +6
We propose a method for metric-scale monocular depth estimation. Inferring depth from a single image is an ill-posed problem due to the loss of scale from perspective projection du…
The Redesigned BaBar Event Store: Believe the Hype
Adeyemi Adesanya, Jacek Becla, Daniel Wang
As the BaBar experiment progresses, it produces new and unforeseen requirements and increasing demands on capacity and feature base. The current system is being utilized well beyon…
Toward Sustainable HPC: Carbon Footprint Estimation and Environmental Implications of HPC Systems
Baolin Li, Rohan Basu Roy, Daniel Wang +3
The rapid growth in demand for HPC systems has led to a rise in carbon footprint, which requires urgent intervention. In this work, we present a comprehensive analysis of the carbo…
A Microservice-Based Platform for Sustainable and Intelligent SLO Fulfilment and Service Management
Juan Luis Herrera, Daniel Wang, Schahram Dustdar
The Microservices Architecture (MSA) design pattern has become a staple for modern applications, allowing functionalities to be divided across fine-grained microservices, fostering…
Iris: Integrating Language into Diffusion-based Monocular Depth Estimation
Ziyao Zeng, Jingcheng Ni, Daniel Wang +5
Traditional monocular depth estimation suffers from inherent ambiguity and visual nuisances. We demonstrate that language can enhance monocular depth estimation by providing an add…
On the Viability of Monocular Depth Pre-training for Semantic Segmentation
Dong Lao, Fengyu Yang, Daniel Wang +4
The question of whether pre-training on geometric tasks is viable for downstream transfer to semantic tasks is important for two reasons, one practical and the other scientific. If…
Periods and atomic firing sequences of parallel chip-firing games on directed graphs
David Ji, Michael Li, Daniel Wang
In 1992, Bitar and Goles introduced the parallel chip-firing game on undirected graphs. Two years later, Prisner extended the game to directed graphs. While the properties of paral…
Binding Touch to Everything: Learning Unified Multimodal Tactile Representations
Fengyu Yang, Chao Feng, Ziyang Chen +8
The ability to associate touch with other modalities has huge implications for humans and computational systems. However, multimodal learning with touch remains challenging due to…
Fine Tuning Methods for Low-resource Languages
Tim Bakkenes, Daniel Wang, Anton Johansson
The rise of Large Language Models has not been inclusive of all cultures. The models are mostly trained on English texts and culture which makes them underperform in other language…
VRSight: An AI-Driven Scene Description System to Improve Virtual Reality Accessibility for Blind People
Daniel Killough, Justin Feng, Zheng Xue "ZX" Ching +4
Virtual Reality (VR) is inaccessible to blind people. While research has investigated many techniques to enhance VR accessibility, they require additional developer effort to integ…
On the Verge of One Petabyte - the Story Behind the BaBar Database System
Adeyemi Adesanya, Tofigh Azemoon, Jacek Becla +8
The BaBar database has pioneered the use of a commercial ODBMS within the HEP community. The unique object-oriented architecture of Objectivity/DB has made it possible to manage ov…
Measuring Stereotype and Deviation Biases in Large Language Models
Daniel Wang, Eli Brignac, Minjia Mao +1
Large language models (LLMs) are widely applied across diverse domains, raising concerns about their limitations and potential risks. In this study, we investigate two types of bia…
Paper 1: The JWST PEARLS View of the El Gordo Galaxy Cluster and of the Structure It Magnifies
Brenda L. Frye, Massimo Pascale, Nicholas Foo +41
The massive galaxy cluster El Gordo (z=0.87) imprints multitudes of gravitationally lensed arcs onto James Webb Space Telescope (JWST) Near-Infrared Camera (NIRCam) images. Eight b…
Hiding behind a curtain of dust: Gas and dust properties of an ultra-luminous strongly-lensed z = 3.75 galaxy behind the Milky Way disk
Belén Alcalde Pampliega, Kevin C. Harrington, Aristeidis Amvrosiadis +21
We present a detailed analysis of J154506, a strongly lensed submillimeter galaxy behind the Lupus-I molecular cloud, and characterisation of its physical properties using a combin…
The Hot and Energetic Universe: A White Paper presenting the science theme motivating the Athena+ mission
Kirpal Nandra, Didier Barret, Xavier Barcons +237
This White Paper, submitted to the recent ESA call for science themes to define its future large missions, advocates the need for a transformational leap in our understanding of tw…
Lomekwi: Resource-Bounded Tool Discovery in LLM Agents
Roshan Klein-Seetharaman, Daniel Wang, Andrew Xu
Existing tool-use benchmarks report a single success rate for complex, multistep tasks. Inspired by ideas from cognitive science, we distinguish tool use from tool discovery and de…
Comparing Fairness of Generative Mobility Models
Daniel Wang, Jack McFarland, Afra Mashhadi +1
This work examines the fairness of generative mobility models, addressing the often overlooked dimension of equity in model performance across geographic regions. Predictive models…
Non-Stabilizing Parallel Chip-Firing Games
David Ji, Michael Li, Daniel Wang
In 2010, Kominers and Kominers proved that any parallel chip-firing game on with chips stabilizes. Recently, Bu, Choi, and Xu made the bound exact: a…
Hybrid Brain-Machine Interface: Integrating EEG and EMG for Reduced Physical Demand
Daniel Wang, Katie Hong, Zachary Sayyah +6
We present a hybrid brain-machine interface (BMI) that integrates steady-state visually evoked potential (SSVEP)-based EEG and facial EMG to improve multimodal control and mitigate…
Beyond ReinMax: Low-Variance Gradient Estimators for Discrete Latent Variables
Daniel Wang, Thang D. Bui
Machine learning models involving discrete latent variables require gradient estimators to facilitate backpropagation in a computationally efficient manner. The most recent additio…
All parallel chip-firing games with have period
Daniel Wang, Nathan Lannan
In 2010, Levine found that the activity functions of parallel chip-firing games on the complete graph converge to a devil's staircase pattern: adding chips causes activity to…
AllocBench: Measuring Online Tool Allocation Capability in LLM Agents
Daniel Wang, Andrew Xu
Creating a reusable tool is an investment: an agent pays a fixed cost now in exchange for the potential of future reuse. Therefore, a user should prefer an agent that creates a sma…
Biased-Noise Quantum Reed-Solomon Codes and a Tornado Concatenation for Cat Qubits
Cheng-You Ho, Justin Luo, Henry Ng +2
The paper presents a Clifford-only quantum Reed‑Solomon code tailored for biased‑noise cat qubits and introduces a two‑layer “Tornado” concatenation that combines an outer RS code…
Divide and Conquer: Text Semantic Matching with Disentangled Keywords and Intents
Yicheng Zou, Hongwei Liu, Tao Gui +5
Text semantic matching is a fundamental task that has been widely used in various scenarios, such as community question answering, information retrieval, and recommendation. Most s…
WorDepth: Variational Language Prior for Monocular Depth Estimation
Ziyao Zeng, Daniel Wang, Fengyu Yang +6
Three-dimensional (3D) reconstruction from a single image is an ill-posed problem with inherent ambiguities, i.e. scale. Predicting a 3D scene from text description(s) is similarly…
Simulated Power of Some Discrete Goodness-of-Fit Test Statistics For Testing the Null Hypothesis of a Zig-Zag Distribution
Clement Ampadu, Daniel Wang, Michael Steele
In this paper, we compare the powers of several discrete goodness-of-fit test statistics considered by Steele and Chaseling [10] under the null hypothesis of a 'zig-zag' distributi…