arXiv subjects

Hang Yang

Publications and source records attributed to Hang Yang.

At least 55 records · Page 3Linked to original sources

arXiv2025

Spec-Driven AI for Science: The ARIA Framework for Automated and Reproducible Data Analysis

The rapid expansion of scientific data has widened the gap between analytical capability and research intent. Existing AI-based analysis tools, ranging from AutoML frameworks to agentic research assistants, either favor automation over transparency or depend on manual scripting that hinders scalability and reproducibility. We present ARIA (Automated Research Intelligence Assistant), a spec-driven, human-in-the-loop framework for automated and interpretable data analysis. ARIA integrates six interoperable layers, namely Command, Context, Code, Data, Orchestration, and AI Module, within a document-centric workflow that unifies human reasoning and machine execution. Through natural-language specifications, researchers define analytical goals while ARIA autonomously generates executable code, validates computations, and produces transparent documentation. Beyond achieving high predictive accuracy, ARIA can rapidly identify optimal feature sets and select suitable models, minimizing redundant tuning and repetitive experimentation. In the Boston Housing case, ARIA discovered 25 key features and determined XGBoost as the best performing model (R square = 0.93) with minimal overfitting. Evaluations across heterogeneous domains demonstrate ARIA's strong performance, interpretability, and efficiency compared with state-of-the-art systems. By combining AI for research and AI for science principles within a spec-driven architecture, ARIA establishes a new paradigm for transparent, collaborative, and reproducible scientific discovery.

cs.AI↗

arXiv2025

Heat Transfer Shape Optimization: Stability and Non-Optimality of the Ball

This paper investigates shape optimization problems in the context of heat transfer, with a focus on the stability and non-optimality of round domains under Robin boundary conditions. Using the flow approach and Steklov eigenvalue estimates, we derive the necessary and sufficient stability conditions for a ball to maximize the averaged heat when the heat source is radially decreasing. Our results show that, counterintuitively, a ball may not be optimal for maximizing the averaged heat under heat convection, even with radially decreasing heat sources located on the center of the ball. Moreover, we identify stability-breaking phenomena by giving precise values of thresholds, which depend on the Robin coefficient, dimension, and volume constraints. Additionally, we demonstrate that a ball can maximize the averaged temperature under certain conditions and we also explore optimal shapes in thin insulation problems.

math.AP↗

arXiv2025

GalaxyGenius: Mock galaxy image generator for various telescopes from hydrodynamical simulations

We introduce GalaxyGenius, a Python package designed to produce synthetic galaxy images tailored to different telescopes based on hydrodynamical simulations. Its implementation will support and advance research on galaxies in the era of large-scale sky surveys. The package comprises three main modules: data preprocessing, ideal data cube generation, and mock observation. Specifically, the preprocessing module extracts necessary properties of star and gas particles for a selected subhalo from hydrodynamical simulations and creates the execution file for the following radiative transfer procedure. Subsequently, building on the above information, the ideal data cube generation module executes a widely used radiative transfer project, specifically the SKIRT, to perform the SED assignment for each particle and the radiative transfer procedure to produce an IFU-like ideal data cube. Lastly, the mock observation module takes the ideal data cube and applies the throughputs of aiming telescopes, while also incorporating the relevant instrumental effects, point spread functions (PSFs), and background noise to generate the required mock observational images of galaxies. To showcase the outcomes of GalaxyGenius, we created a series of mock images of galaxies based on the IllustrisTNG and EAGLE simulations for both space and ground-based surveys, spanning ultraviolet (UV) to infrared (IR) wavelength coverage, including CSST, Euclid, HST, JWST, Roman, and HSC. GalaxyGenius offers a flexible framework to generate mock galaxy images with customizable recipes. These generated images can serve as valuable references for verifying and validating new approaches in astronomical research. They can also serve as training sets for relevant studies using deep learning in cases where real observational data are insufficient.

astro-ph.IM↗

arXiv2025

Pieceformer: Similarity-Driven Knowledge Transfer via Scalable Graph Transformer in VLSI

Accurate graph similarity is critical for knowledge transfer in VLSI design, enabling the reuse of prior solutions to reduce engineering effort and turnaround time. We propose Pieceformer, a scalable, self-supervised similarity assessment framework, equipped with a hybrid message-passing and graph transformer encoder. To address transformer scalability, we incorporate a linear transformer backbone and introduce a partitioned training pipeline for efficient memory and parallelism management. Evaluations on synthetic and real-world CircuitNet datasets show that Pieceformer reduces mean absolute error (MAE) by 24.9% over the baseline and is the only method to correctly cluster all real-world design groups. We further demonstrate the practical usage of our model through a case study on a partitioning task, achieving up to 89% runtime reduction. These results validate the framework's effectiveness for scalable, unbiased design reuse in modern VLSI systems.

cs.LG↗

arXiv2025

Learning Class Prototypes for Unified Sparse Supervised 3D Object Detection

Both indoor and outdoor scene perceptions are essential for embodied intelligence. However, current sparse supervised 3D object detection methods focus solely on outdoor scenes without considering indoor settings. To this end, we propose a unified sparse supervised 3D object detection method for both indoor and outdoor scenes through learning class prototypes to effectively utilize unlabeled objects. Specifically, we first propose a prototype-based object mining module that converts the unlabeled object mining into a matching problem between class prototypes and unlabeled features. By using optimal transport matching results, we assign prototype labels to high-confidence features, thereby achieving the mining of unlabeled objects. We then present a multi-label cooperative refinement module to effectively recover missed detections through pseudo label quality control and prototype label cooperation. Experiments show that our method achieves state-of-the-art performance under the one object per scene sparse supervised setting across indoor and outdoor datasets. With only one labeled object per scene, our method achieves about 78%, 90%, and 96% performance compared to the fully supervised detector on ScanNet V2, SUN RGB-D, and KITTI, respectively, highlighting the scalability of our method. Code is available at https://github.com/zyrant/CPDet3D.

cs.CV↗

arXiv2025

Flow approach on Riesz type nonlocal energies

Via continuous deformations based on natural flow evolutions, we prove several novel monotonicity results for Riesz-type nonlocal energies on triangles and quadrilaterals. Some of these results imply new and simpler proofs for known theorems without relying on any symmetrization arguments.

math.AP↗

arXiv2025

Modeling the cool gas clumps in the circumgalactic medium

A major challenge in CGM studies is determining the three-dimensional (3D) properties from the observed projected observations. Here, we decompose the 3D gas density and spatial distribution of cool clouds by fitting a cool CGM model with the absorption observations, including the cool gas density, Ly$α$, and Mg II equivalent widths. The clumpiness in the cool CGM is considered by modeling individual clouds. This model has four major components: the radial profile of the cool gas density; the number density of clouds; the absorption properties within individual clouds; and the velocity dispersion in the CGM. The observed cool gas density exhibits a large dispersion of $\approx2-3$ dex within the virial radius ($r_{vir}$). This dispersion can be reproduced with a combination of the projection effect (i.e., distant low-density clouds projected at small radii) and the intrinsic variation in the gas density. By modeling the probability density functions of gas density at different radii, the cool gas density is modeled as a $β$-model with a characteristic gas density of $\log n_{H,0}/{\rm cm^{-3}}=-2.57_{-0.25}^{+0.43}$ at $r_{vir}$ and a slope of $β_c=0.63_{-0.20}^{+0.16}$, and the intrinsic dispersion is $σ_{n_{H}}\approx 0.56_{-0.20}^{+0.19}$ dex. Assuming a cloud mass of $10^4M_\odot$, we further constrain the number density of cool clouds by jointly reproducing Ly$α$ and Mg II equivalent width samples, resulting into a number density of $\log n_{\mathcal{N}_{cl},0}/ r_{vir}^{-3}=4.76^{+0.27}_{-0.21}$ at $r_{vir}$ and a slope of $β_N=0.65^{+0.06}_{-0.07}$. This spatial distribution of the cool CGM leads to a total cool gas mass of $\log M_{cool}/M_\odot=10.01^{+0.06}_{-0.06}$ for $L^*$ galaxies, while varying the cloud mass from $10^3M_\odot$ to $10^6M_\odot$ leads to the total cool CGM mass of $9.62_{-0.07}^{+0.05}$ to $10.46_{-0.05}^{+0.05}$.

astro-ph.GA↗

arXiv2025

PTQ4RIS: Post-Training Quantization for Referring Image Segmentation

Referring Image Segmentation (RIS), aims to segment the object referred by a given sentence in an image by understanding both visual and linguistic information. However, existing RIS methods tend to explore top-performance models, disregarding considerations for practical applications on resources-limited edge devices. This oversight poses a significant challenge for on-device RIS inference. To this end, we propose an effective and efficient post-training quantization framework termed PTQ4RIS. Specifically, we first conduct an in-depth analysis of the root causes of performance degradation in RIS model quantization and propose dual-region quantization (DRQ) and reorder-based outlier-retained quantization (RORQ) to address the quantization difficulties in visual and text encoders. Extensive experiments on three benchmarks with different bits settings (from 8 to 4 bits) demonstrates its superior performance. Importantly, we are the first PTQ method specifically designed for the RIS task, highlighting the feasibility of PTQ in RIS applications. Code and video are available at {https://github.com/gugu511yy/PTQ4RIS}.

cs.CV↗

arXiv2025

ChallengeMe: An Adversarial Learning-enabled Text Summarization Framework

The astonishing performance of large language models (LLMs) and their remarkable achievements in production and daily life have led to their widespread application in collaborative tasks. However, current large models face challenges such as hallucination and lack of specificity in content generation in vertical domain tasks. Inspired by the contrast and classification mechanisms in human cognitive processes, this paper constructs an adversarial learning-based prompt framework named ChallengeMe, which includes three cascaded solutions: generation prompts, evaluation prompts, and feedback optimization. In this process, we designed seven core optimization dimensions and set the threshold for adversarial learning. The results of mixed case studies on the text summarization task show that the proposed framework can generate more accurate and fluent text summaries compared to the current advanced mainstream LLMs.

cs.CL↗

arXiv2025

LLM-MedQA: Enhancing Medical Question Answering through Case Studies in Large Language Models

Accurate and efficient question-answering systems are essential for delivering high-quality patient care in the medical field. While Large Language Models (LLMs) have made remarkable strides across various domains, they continue to face significant challenges in medical question answering, particularly in understanding domain-specific terminologies and performing complex reasoning. These limitations undermine their effectiveness in critical medical applications. To address these issues, we propose a novel approach incorporating similar case generation within a multi-agent medical question-answering (MedQA) system. Specifically, we leverage the Llama3.1:70B model, a state-of-the-art LLM, in a multi-agent architecture to enhance performance on the MedQA dataset using zero-shot learning. Our method capitalizes on the model's inherent medical knowledge and reasoning capabilities, eliminating the need for additional training data. Experimental results show substantial performance gains over existing benchmark models, with improvements of 7% in both accuracy and F1-score across various medical QA tasks. Furthermore, we examine the model's interpretability and reliability in addressing complex medical queries. This research not only offers a robust solution for medical question answering but also establishes a foundation for broader applications of LLMs in the medical domain.

cs.CL↗

arXiv2024

TDCNet: Transparent Objects Depth Completion with CNN-Transformer Dual-Branch Parallel Network

The sensing and manipulation of transparent objects present a critical challenge in industrial and laboratory robotics. Conventional sensors face challenges in obtaining the full depth of transparent objects due to the refraction and reflection of light on their surfaces and their lack of visible texture. Previous research has attempted to obtain complete depth maps of transparent objects from RGB and damaged depth maps (collected by depth sensor) using deep learning models. However, existing methods fail to fully utilize the original depth map, resulting in limited accuracy for deep completion. To solve this problem, we propose TDCNet, a novel dual-branch CNN-Transformer parallel network for transparent object depth completion. The proposed framework consists of two different branches: one extracts features from partial depth maps, while the other processes RGB-D images. Experimental results demonstrate that our model achieves state-of-the-art performance across multiple public datasets. Our code and the pre-trained model are publicly available at https://github.com/XianghuiFan/TDCNet.

cs.CV↗

arXiv2024

Apostle--Auriga: Effects of stellar feedback subgrid models on the evolution of angular momentum in disc galaxies

Utilizing the Apostle--Auriga simulations, which start from the same zoom-in initial conditions of Local Group-like systems but run with different galaxy formation subgrid models and hydrodynamic solvers, we study the impact of stellar feedback models on the evolution of angular momentum in disc galaxies. At $z = 0$, Auriga disc galaxies tend to exhibit higher specific angular momenta compared to their cross-matched Apostle counterparts. By tracing the evolution history of the Lagrangian mass tracers of the in-situ star particles in the $z = 0$ galaxies, we find that the specific angular momentum distributions of the gas tracers from the two simulations at the halo accretion time are relatively similar. The present-day angular momentum difference is mainly driven by the physical processes occurring inside dark matter haloes, especially galactic fountains. Due to the different subgrid implementations of stellar feedback processes, Auriga galaxies contain a high fraction of gas that has gone through recycled fountain (${\sim} 65$ per cent) which could acquire angular momentum through mixing with the high angular momentum circumgalactic medium (CGM). In Apostle, however, the fraction of gas that has undergone the recycled fountain process is significantly lower (down to ${\sim} 20$ per cent for Milky Way-sized galaxies) and the angular momentum acquisition from the CGM is marginal. As a result, the present-day Auriga galaxies overall have higher specific angular momenta.

astro-ph.GA↗

arXiv2024

The formation and survival of the Milky Way's oldest stellar disk

It remains a mystery when our Milky Way first formed a stellar disk component that survived and maintained its disk structure from subsequent galaxy mergers. We present a study of the age-dependent structure and star formation rate of the Milky Way's disk using high-alpha stars with significant orbital angular momentum that have precise age determinations. Our results show that the radial scale length is nearly independent of age, while the vertical scale height experienced dramatic evolution. A disk-like geometry presents even for populations older than 13 Gyr, with the scale height-to-length ratio dropping below 0.5 for populations younger than 12.5 Gyr. We dub the oldest population that has maintained a disk geometry - apparently formed over 13 Gyr ago - PanGu. With an estimated present-day stellar mass of $2 \times 10^9$ $M_\odot$, PanGu is presumed to be a major stellar component of our Galaxy in the earliest epoch. The total present-day stellar mass of the whole high-alpha disk is $2 \times 10^{10}$ $M_\odot$, mostly formed during a distinct star formation rate peak of 11 $M_\odot$ per year around 11 Gyrs ago. A comparison with Milky Way analogs in the TNG50 simulations implies that our Galaxy has experienced an exceptionally quiescent dynamical history, even before the Gaia-Enceladus merger.

astro-ph.GA↗

arXiv2024

ERCache: An Efficient and Reliable Caching Framework for Large-Scale User Representations in Meta's Ads System

The increasing complexity of deep learning models used for calculating user representations presents significant challenges, particularly with limited computational resources and strict service-level agreements (SLAs). Previous research efforts have focused on optimizing model inference but have overlooked a critical question: is it necessary to perform user model inference for every ad request in large-scale social networks? To address this question and these challenges, we first analyze user access patterns at Meta and find that most user model inferences occur within a short timeframe. T his observation reveals a triangular relationship among model complexity, embedding freshness, and service SLAs. Building on this insight, we designed, implemented, and evaluated ERCache, an efficient and robust caching framework for large-scale user representations in ads recommendation systems on social networks. ERCache categorizes cache into direct and failover types and applies customized settings and eviction policies for each model, effectively balancing model complexity, embedding freshness, and service SLAs, even considering the staleness introduced by caching. ERCache has been deployed at Meta for over six months, supporting more than 30 ranking models while efficiently conserving computational resources and complying with service SLA requirements.

cs.IR↗

arXiv2024

Optimization of Nuclear Mass Models Using Algorithms and Neural Networks

Taking into account nucleon-nucleon gravitational interaction, higher-order terms of symmetry energy, pairing interaction, and neural network corrections, a new BW4 mass model has been developed, which more accurately reflects the contributions of various terms to the binding energy. A novel hybrid algorithm and neural network correction method has been implemented to optimize the discrepancy between theoretical and experimental results, significantly improving the model's binding energy predictions (reduced to around 350 keV). At the same time, the theoretical accuracy near magic nuclei has been marginally enhanced, effectively capturing the special interaction effects around magic nuclei and showing good agreement with experimental data.

nucl-th↗

arXiv2024

Kinematic morphology of low-mass galaxies in IllustrisTNG

The origin of diverse kinematic morphologies observed in low-mass galaxies is unclear. In this study, we investigate the kinematic morphologies of central galaxies with stellar mass $10^{8.5-9.0} M_\odot$ at $z=0$ in the TNG50-1 cosmological simulation. The majority of the low-mass galaxies in TNG50-1 are dispersion-dominated, consistent with observations. By tracing the evolutionary histories of simulated low-mass galaxies, we find that while most stars form in rotating cold gas discs, the orientation of the star-forming discs relative to the galaxies may evolve with cosmic time. If the cold gas disc remains aligning with the galaxy during its evolution, stars formed at different times share the same rotational direction, leading to a rotation-dominated system. On the contrary, frequent misalignment of cold gas disc would result in a dispersion-dominated system. In addition, we also find that the two-body scattering can have a non-negligible numerical heating effect on the simulated galaxy morphology, especially at central regions of galaxies and for relatively low-mass galaxies. By comparing results of simulations with different resolutions, our results suggest that the simulated morphology of galaxies is roughly reliable when their number of stellar particles exceeds about $10^{4}$, and bulge morphology of galaxies can not be resolved robustly at the resolution level of TNG50-1.

astro-ph.GA↗

arXiv2024

Empowering Working Memory for Large Language Model Agents

Large language models (LLMs) have achieved impressive linguistic capabilities. However, a key limitation persists in their lack of human-like memory faculties. LLMs exhibit constrained memory retention across sequential interactions, hindering complex reasoning. This paper explores the potential of applying cognitive psychology's working memory frameworks, to enhance LLM architecture. The limitations of traditional LLM memory designs are analyzed, including their isolation of distinct dialog episodes and lack of persistent memory links. To address this, an innovative model is proposed incorporating a centralized Working Memory Hub and Episodic Buffer access to retain memories across episodes. This architecture aims to provide greater continuity for nuanced contextual reasoning during intricate tasks and collaborative scenarios. While promising, further research is required into optimizing episodic memory encoding, storage, prioritization, retrieval, and security. Overall, this paper provides a strategic blueprint for developing LLM agents with more sophisticated, human-like memory capabilities, highlighting memory mechanisms as a vital frontier in artificial general intelligence.

cs.CL↗

arXiv2024

A Method for Parsing and Vectorization of Semi-structured Data used in Retrieval Augmented Generation

This paper presents a novel method for parsing and vectorizing semi-structured data to enhance the functionality of Retrieval-Augmented Generation (RAG) within Large Language Models (LLMs). We developed a comprehensive pipeline for converting various data formats into .docx, enabling efficient parsing and structured data extraction. The core of our methodology involves the construction of a vector database using Pinecone, which integrates seamlessly with LLMs to provide accurate, context-specific responses, particularly in environmental management and wastewater treatment operations. Through rigorous testing with both English and Chinese texts in diverse document formats, our results demonstrate a marked improvement in the precision and reliability of LLMs outputs. The RAG-enhanced models displayed enhanced ability to generate contextually rich and technically accurate responses, underscoring the potential of vector knowledge bases in significantly boosting the performance of LLMs in specialized domains. This research not only illustrates the effectiveness of our method but also highlights its potential to revolutionize data processing and analysis in environmental sciences, setting a precedent for future advancements in AI-driven applications. Our code is available at https://github.com/linancn/TianGong-AI-Unstructure.git.

cs.DB↗