arXiv subjects

Jie Jiang

Publications and source records attributed to Jie Jiang.

At least 109 records · Page 6Linked to original sources

arXiv2025

Probing Solar Polar Regions

The magnetic fields and dynamical processes in the solar polar regions play a crucial role in the solar magnetic cycle and in supplying mass and energy to the fast solar wind, ultimately being vital in controlling solar activities and driving space weather. Despite numerous efforts to explore these regions, to date no imaging observations of the Sun's poles have been achieved from vantage points out of the ecliptic plane, leaving their behavior and evolution poorly understood. This observation gap has left three top-level scientific questions unanswered, 1) How does the solar dynamo work and drive the solar magnetic cycle? 2) What drives the fast solar wind? 3) How do space weather processes globally originate from the Sun and propagate throughout the solar system? The Solar Polar-orbit Observatory (SPO) mission, a solar polar exploration spacecraft, is proposed to address these three unanswered scientific questions by imaging the Sun's poles from high heliolatitudes. In order to achieve its scientific goals, SPO will carry six remote-sensing and four in-situ instruments to measure the vector magnetic fields and Doppler velocity fields in the photosphere, to observed the Sun in the extreme ultraviolet, X-ray, and radio wavelengths, to image the corona and the heliosphere up to 45 $R_\odot$, and to perform in-situ detection of magnetic fields, and low- and high-energy particles in the solar wind.

astro-ph.SR↗

arXiv2025

Hunyuan3D Studio: End-to-End AI Pipeline for Game-Ready 3D Asset Generation

The creation of high-quality 3D assets, a cornerstone of modern game development, has long been characterized by labor-intensive and specialized workflows. This paper presents Hunyuan3D Studio, an end-to-end AI-powered content creation platform designed to revolutionize the game production pipeline by automating and streamlining the generation of game-ready 3D assets. At its core, Hunyuan3D Studio integrates a suite of advanced neural modules (such as Part-level 3D Generation, Polygon Generation, Semantic UV, etc.) into a cohesive and user-friendly system. This unified framework allows for the rapid transformation of a single concept image or textual description into a fully-realized, production-quality 3D model complete with optimized geometry and high-fidelity PBR textures. We demonstrate that assets generated by Hunyuan3D Studio are not only visually compelling but also adhere to the stringent technical requirements of contemporary game engines, significantly reducing iteration time and lowering the barrier to entry for 3D content creation. By providing a seamless bridge from creative intent to technical asset, Hunyuan3D Studio represents a significant leap forward for AI-assisted workflows in game development and interactive media.

cs.CV↗

arXiv2025

SQLGovernor: An LLM-powered SQL Toolkit for Real World Application

SQL queries in real world analytical environments, whether written by humans or generated automatically often suffer from syntax errors, inefficiency, or semantic misalignment, especially in complex OLAP scenarios. To address these challenges, we propose SQLGovernor, an LLM powered SQL toolkit that unifies multiple functionalities, including syntax correction, query rewriting, query modification, and consistency verification within a structured framework enhanced by knowledge management. SQLGovernor introduces a fragment wise processing strategy to enable fine grained rewriting and localized error correction, significantly reducing the cognitive load on the LLM. It further incorporates a hybrid self learning mechanism guided by expert feedback, allowing the system to continuously improve through DBMS output analysis and rule validation. Experiments on benchmarks such as BIRD and BIRD CRITIC, as well as industrial datasets, show that SQLGovernor consistently boosts the performance of base models by up to 10%, while minimizing reliance on manual expertise. Deployed in production environments, SQLGovernor demonstrates strong practical utility and effective performance.

cs.DB↗

arXiv2025

Simulation of Solar Surface Flux Transport Constrained by Magnetic Power Spectra. I. Flux Transport Parameter

The multi-scale structure of the solar surface magnetic field is essential for understanding both the Sun's internal dynamo processes and its external magnetic activity. The surface flux transport (SFT) model has been successful in describing the large-scale evolution of the surface field, but its ability to capture observed multi-scale features, quantified by magnetic power spectra, remains uncertain. Here, we evaluate the SFT model by comparing observed and simulated power spectra across a broad range of spatial scales and by analyzing the effects of key transport parameters. We find that the simulations reproduce the observed spectra well at spherical harmonic degrees $l\lesssim60$, but diverge progressively at smaller spatial scales $l\gtrsim60$. This divergence likely arises from the diffusion approximation used to model the random walk of supergranulation. Power at $20 \lesssim l \lesssim 60$ is primarily determined by the magnetic flux sources, while at $l \lesssim 20$, the spectra are more sensitive to transport parameters. The meridional flow profile, including its equatorial gradient, peak latitude, and polar distribution, along with turbulent diffusivity, has distinct impacts on the low-degree modes ($l \leq 5$). In particular, a comparison of the $l$=3 and $l$=5 multipoles strengths suggests that the poleward flow above $\sim\pm60^\circ$ latitudes is very weak. This study presents the first quantitative validation of SFT models using magnetic power spectra and provides new constraints on surface flux transport physics.

astro-ph.SR↗

arXiv2025

R-4B: Incentivizing General-Purpose Auto-Thinking Capability in MLLMs via Bi-Mode Annealing and Reinforce Learning

Multimodal Large Language Models (MLLMs) equipped with step-by-step thinking capabilities have demonstrated remarkable performance on complex reasoning problems. However, this thinking process is redundant for simple problems solvable without complex reasoning. To address this inefficiency, we propose R-4B, an auto-thinking MLLM, which can adaptively decide when to think based on problem complexity. The central idea of R-4B is to empower the model with both thinking and non-thinking capabilities using bi-mode annealing, and apply Bi-mode Policy Optimization (BPO) to improve the model's accuracy in determining whether to activate the thinking process. Specifically, we first train the model on a carefully curated dataset spanning various topics, which contains samples from both thinking and non-thinking modes. Then it undergoes a second phase of training under an improved GRPO framework, where the policy model is forced to generate responses from both modes for each input query. Experimental results show that R-4B achieves state-of-the-art performance across 25 challenging benchmarks. It outperforms Qwen2.5-VL-7B in most tasks and achieves performance comparable to larger models such as Kimi-VL-A3B-Thinking-2506 (16B) on reasoning-intensive benchmarks with lower computational cost.

cs.CV↗

arXiv2025

Empowering Large Language Model for Sequential Recommendation via Multimodal Embeddings and Semantic IDs

Sequential recommendation (SR) aims to capture users' dynamic interests and sequential patterns based on their historical interactions. Recently, the powerful capabilities of large language models (LLMs) have driven their adoption in SR. However, we identify two critical challenges in existing LLM-based SR methods: 1) embedding collapse when incorporating pre-trained collaborative embeddings and 2) catastrophic forgetting of quantized embeddings when utilizing semantic IDs. These issues dampen the model scalability and lead to suboptimal recommendation performance. Therefore, based on LLMs like Llama3-8B-instruct, we introduce a novel SR framework named MME-SID, which integrates multimodal embeddings and quantized embeddings to mitigate embedding collapse. Additionally, we propose a Multimodal Residual Quantized Variational Autoencoder (MM-RQ-VAE) with maximum mean discrepancy as the reconstruction loss and contrastive learning for alignment, which effectively preserve intra-modal distance information and capture inter-modal correlations, respectively. To further alleviate catastrophic forgetting, we initialize the model with the trained multimodal code embeddings. Finally, we fine-tune the LLM efficiently using LoRA in a multimodal frequency-aware fusion manner. Extensive experiments on three public datasets validate the superior performance of MME-SID thanks to its capability to mitigate embedding collapse and catastrophic forgetting. The implementation code and datasets are publicly available for reproduction: https://github.com/Applied-Machine-Learning-Lab/MME-SID.

cs.IR↗

arXiv2025

BrainGPT: Unleashing the Potential of EEG Generalist Foundation Model by Autoregressive Pre-training

Electroencephalogram (EEG) signals are pivotal in providing insights into spontaneous brain activity, highlighting their significant importance in neuroscience research. However, the exploration of versatile EEG models is constrained by diverse data formats, outdated pre-training paradigms, and limited transfer learning methods, only leading to specialist models on single dataset. In this paper, we introduce EEGPT, the first generalist EEG foundation model designed to address these challenges. First, we propose an electrode-wise modeling strategy that treats each electrode as a fundamental unit, enabling the integration of diverse EEG datasets collected from up to 138 electrodes, amassing 37.5M pre-training samples. Second, we develop the first autoregressive EEG pre-trained model, moving away from traditional masked autoencoder approaches to a next signal prediction task that better captures the sequential and temporal dependencies of EEG data. We also explore scaling laws with model up to 1.1B parameters: the largest in EEG research to date. Third, we introduce a multi-task transfer learning paradigm using a learnable electrode graph network shared across tasks, which for the first time confirms multi-task compatibility and synergy. As the first generalist EEG foundation model, EEGPT shows broad compatibility with various signal acquisition devices, subjects, and tasks. It supports up to 138 electrodes and any combination thereof as input. Furthermore, we simultaneously evaluate it on 5 distinct tasks across 12 benchmarks. EEGPT consistently outperforms existing specialist models across all downstream tasks, with its effectiveness further validated through extensive ablation studies. This work sets a new direction for generalist EEG modeling, offering improved scalability, transferability, and adaptability for a wide range of EEG applications. The code and models will be released.

eess.SP↗

arXiv2025

Strong Raman Optical Activity and Chiral Phonons in Chiral Hybrid Organic-Inorganic Perovskites

Hybrid organic-inorganic perovskites with chiral organic cations are very interesting for optoelectronic applications because of their intrinsically chiral light-matter interactions. Chiral distortions in these materials lead to circular dichroism, circular birefringence, and circularly polarized luminescence in the band transitions of the inorganic sublattice. Raman-active vibrational modes in these crystals are governed by crystal symmetry and therefore are also strongly impacted by the nature and magnitude of the chiral distortions. Here, we report low-frequency Raman modes that are sensitive to circularly polarized excitation in chiral hybrid organic-inorganic perovskites (CHOIPs) across a wide range of structures and compositions. The circularly polarized Raman spectra from enantiomers of CHOIP single crystals exhibit sharp modes below 150 cm-1, corresponding to vibrations of the lead iodide octahedra. These modes exhibit strong differences in intensities (Raman optical activity, ROA) depending on the handedness of the excitation, with high degree of polarization for several modes. Calculations reveal the presence of several chiral phonon modes with opposite phonon angular momenta. The strong ROA and the chiral phonon modes are a direct consequence of chirality transfer from the chiral organic linker to the lead iodide octahedra in the CHOIP structure, resulting in a strong chiroptical response in the phonon modes.

cond-mat.mtrl-sci↗

arXiv2025

Large Foundation Model for Ads Recommendation

Online advertising relies on accurate recommendation models, with recent advances using pre-trained large-scale foundation models (LFMs) to capture users' general interests across multiple scenarios and tasks. However, existing methods have critical limitations: they extract and transfer only user representations (URs), ignoring valuable item representations (IRs) and user-item cross representations (CRs); and they simply use a UR as a feature in downstream applications, which fails to bridge upstream-downstream gaps and overlooks more transfer granularities. In this paper, we propose LFM4Ads, an All-Representation Multi-Granularity transfer framework for ads recommendation. It first comprehensively transfers URs, IRs, and CRs, i.e., all available representations in the pre-trained foundation model. To effectively utilize the CRs, it identifies the optimal extraction layer and aggregates them into transferable coarse-grained forms. Furthermore, we enhance the transferability via multi-granularity mechanisms: non-linear adapters for feature-level transfer, an Isomorphic Interaction Module for module-level transfer, and Standalone Retrieval for model-level transfer. LFM4Ads has been successfully deployed in Tencent's industrial-scale advertising platform, processing tens of billions of daily samples while maintaining terabyte-scale model parameters with billions of sparse embedding keys across approximately two thousand features. Since its production deployment in Q4 2024, LFM4Ads has achieved 10+ successful production launches across various advertising scenarios, including primary ones like Weixin Moments and Channels. These launches achieve an overall GMV lift of 2.45% across the entire platform, translating to estimated annual revenue increases in the hundreds of millions of dollars.

cs.LG↗

arXiv2025

HunyuanWorld 1.0: Generating Immersive, Explorable, and Interactive 3D Worlds from Words or Pixels

Creating immersive and playable 3D worlds from texts or images remains a fundamental challenge in computer vision and graphics. Existing world generation approaches typically fall into two categories: video-based methods that offer rich diversity but lack 3D consistency and rendering efficiency, and 3D-based methods that provide geometric consistency but struggle with limited training data and memory-inefficient representations. To address these limitations, we present HunyuanWorld 1.0, a novel framework that combines the best of both worlds for generating immersive, explorable, and interactive 3D scenes from text and image conditions. Our approach features three key advantages: 1) 360° immersive experiences via panoramic world proxies; 2) mesh export capabilities for seamless compatibility with existing computer graphics pipelines; 3) disentangled object representations for augmented interactivity. The core of our framework is a semantically layered 3D mesh representation that leverages panoramic images as 360° world proxies for semantic-aware world decomposition and reconstruction, enabling the generation of diverse 3D worlds. Extensive experiments demonstrate that our method achieves state-of-the-art performance in generating coherent, explorable, and interactive 3D worlds while enabling versatile applications in virtual reality, physical simulation, game development, and interactive content creation.

cs.CV↗

arXiv2025

MaterialMVP: Illumination-Invariant Material Generation via Multi-view PBR Diffusion

Physically-based rendering (PBR) has become a cornerstone in modern computer graphics, enabling realistic material representation and lighting interactions in 3D scenes. In this paper, we present MaterialMVP, a novel end-to-end model for generating PBR textures from 3D meshes and image prompts, addressing key challenges in multi-view material synthesis. Our approach leverages Reference Attention to extract and encode informative latent from the input reference images, enabling intuitive and controllable texture generation. We also introduce a Consistency-Regularized Training strategy to enforce stability across varying viewpoints and illumination conditions, ensuring illumination-invariant and geometrically consistent results. Additionally, we propose Dual-Channel Material Generation, which separately optimizes albedo and metallic-roughness (MR) textures while maintaining precise spatial alignment with the input images through Multi-Channel Aligned Attention. Learnable material embeddings are further integrated to capture the distinct properties of albedo and MR. Experimental results demonstrate that our model generates PBR textures with realistic behavior across diverse lighting scenarios, outperforming existing methods in both consistency and quality for scalable 3D asset creation.

cs.CV↗

arXiv2025

SLTarch: Towards Scalable Point-Based Neural Rendering by Taming Workload Imbalance and Memory Irregularity

Rendering is critical in fields like 3D modeling, AR/VR, and autonomous driving, where high-quality, real-time output is essential. Point-based neural rendering (PBNR) offers a photorealistic and efficient alternative to conventional methods, yet it is still challenging to achieve real-time rendering on mobile platforms. We pinpoint two major bottlenecks in PBNR pipelines: LoD search and splatting. LoD search suffers from workload imbalance and irregular memory access, making it inefficient on off-the-shelf GPUs. Meanwhile, splatting introduces severe warp divergence across GPU threads due to its inherent sparsity. To tackle these challenges, we propose SLTarch, an algorithm-architecture co-designed framework. At its core, SLTarch introduces SLTree, a dedicated subtree-based data structure, and LTcore, a specialized hardware architecture tailored for efficient LoD search. Additionally, we co-design a divergence-free splatting algorithm with our simple yet principled hardware augmentation, SPcore, to existing PBNR accelerators. Compared to a mobile GPU, SLTarch achieves 3.9$\times$ speedup and 98\% energy savings with negligible architecture overhead. Compared to existing accelerator designs, SLTarch achieves 1.8$\times$ speedup with 54\% energy savings.

cs.AR↗

arXiv2025

X-Omni: Reinforcement Learning Makes Discrete Autoregressive Image Generative Models Great Again

Numerous efforts have been made to extend the ``next token prediction'' paradigm to visual contents, aiming to create a unified approach for both image generation and understanding. Nevertheless, attempts to generate images through autoregressive modeling with discrete tokens have been plagued by issues such as low visual fidelity, distorted outputs, and failure to adhere to complex instructions when rendering intricate details. These shortcomings are likely attributed to cumulative errors during autoregressive inference or information loss incurred during the discretization process. Probably due to this challenge, recent research has increasingly shifted toward jointly training image generation with diffusion objectives and language generation with autoregressive objectives, moving away from unified modeling approaches. In this work, we demonstrate that reinforcement learning can effectively mitigate artifacts and largely enhance the generation quality of a discrete autoregressive modeling method, thereby enabling seamless integration of image and language generation. Our framework comprises a semantic image tokenizer, a unified autoregressive model for both language and images, and an offline diffusion decoder for image generation, termed X-Omni. X-Omni achieves state-of-the-art performance in image generation tasks using a 7B language model, producing images with high aesthetic quality while exhibiting strong capabilities in following instructions and rendering long texts.

cs.CV↗

arXiv2025

SiriusBI: A Comprehensive LLM-Powered Solution for Data Analytics in Business Intelligence

With the proliferation of Large Language Models (LLMs) in Business Intelligence (BI), existing solutions face critical challenges in industrial deployments: functionality deficiencies from legacy systems failing to meet evolving LLM-era user demands, interaction limitations from single-round SQL generation paradigms inadequate for multi-round clarification, and cost for domain adaptation arising from cross-domain methods migration. We present SiriusBI, a practical LLM-powered BI system addressing the challenges of industrial deployments through three key innovations: (a) An end-to-end architecture integrating multi-module coordination to overcome functionality gaps in legacy systems; (b) A multi-round dialogue with querying mechanism, consisting of semantic completion, knowledge-guided clarification, and proactive querying processes, to resolve interaction constraints in SQL generation; (c) A data-conditioned SQL generation method selection strategy that supports both an efficient one-step Fine-Tuning approach and a two-step method leveraging Semantic Intermediate Representation for low-cost cross-domain applications. Experiments on both real-world datasets and public benchmarks demonstrate the effectiveness of SiriusBI. User studies further confirm that SiriusBI enhances both productivity and user experience. As an independent service on Tencent's data platform, SiriusBI is deployed across finance, advertising, and cloud sectors, serving dozens of enterprise clients. It achieves over 93% accuracy in SQL generation and reduces data analysts' query time from minutes to seconds in real-world applications.

cs.DB↗

arXiv2025

LEADRE: Multi-Faceted Knowledge Enhanced LLM Empowered Display Advertisement Recommender System

Display advertising provides significant value to advertisers, publishers, and users. Traditional display advertising systems utilize a multi-stage architecture consisting of retrieval, coarse ranking, and final ranking. However, conventional retrieval methods rely on ID-based learning to rank mechanisms and fail to adequately utilize the content information of ads, which hampers their ability to provide diverse recommendation lists. To address this limitation, we propose leveraging the extensive world knowledge of LLMs. However, three key challenges arise when attempting to maximize the effectiveness of LLMs: "How to capture user interests", "How to bridge the knowledge gap between LLMs and advertising system", and "How to efficiently deploy LLMs". To overcome these challenges, we introduce a novel LLM-based framework called LLM Empowered Display ADvertisement REcommender system (LEADRE). LEADRE consists of three core modules: (1) The Intent-Aware Prompt Engineering introduces multi-faceted knowledge and designs intent-aware pairs that fine-tune LLMs to generate ads tailored to users' personal interests. (2) The Advertising-Specific Knowledge Alignment incorporates auxiliary fine-tuning tasks and Direct Preference Optimization (DPO) to align LLMs with ad semantic and business value. (3) The Efficient System Deployment deploys LEADRE in an online environment by integrating both latency-tolerant and latency-sensitive service. Extensive offline experiments demonstrate the effectiveness of LEADRE and validate the contributions of individual modules. Online A/B test shows that LEADRE leads to a 1.57% and 1.17% GMV lift for serviced users on WeChat Channels and Moments separately. LEADRE has been deployed on both platforms, serving tens of billions of requests each day.

cs.IR↗

arXiv2025

Minimal Roles of Solar Subsurface Meridional Flow in the distributed-shear Babcock-Leighton Dynamo

The subsurface meridional flow has long been recognized as a critical factor in driving the solar cycle. Specifically, the equatorward return flow in the tachocline is widely believed to be responsible for the formation of the sunspot butterfly diagram and determine the solar cycle period within the framework of flux transport dynamo (FTD) models. We aim to investigate whether the subsurface meridional flow also plays a significant role in the recently developed distributed-shear Babcock-Leighton (BL) dynamo model, which operates within the convection zone, rather than the tachocline. Various meridional flow configurations, including a deep single cell, a shallow single cell, and double cells, are applied in the distributed-shear BL dynamo model to explore the mechanisms driving the butterfly diagram and variations in the cycle period. Subsurface meridional flow plays a minimal role in the distributed-shear BL dynamo. A solar-like butterfly diagram can be generated even with a double-cell meridional flow. The diagram arises from the time- and latitude-dependent regeneration of the toroidal field, governed by latitude-dependent latitudinal differential rotation and the evolution of surface magnetic fields. The cycle period is determined by the surface flux source and transport process responsible for polar field generation, which corresponds to the $α$-effect in the BL-type dynamo. The cycle period may exhibit varying dependence on the amplitude of the subsurface flow. The distributed-shear BL dynamo differs fundamentally from the FTD models, as it does not rely on the subsurface flux transport. This distinction aligns the distributed-shear BL dynamo more closely with the original BL dynamo and the conventional $αΩ$ dynamo. Although the subsurface meridional flow plays a negligible role in our distributed-shear BL dynamo, the poleward surface flow is essential.

astro-ph.SR↗

arXiv2025

Fully stabilized Er fiber comb at 1 GHz by harmonic modelocking

Modelocked frequency comb lasers have always operated with a single pulse circulating in the laser cavity. This meant that each laser technology had an associated limit on pulse repetition rate. Achieving higher rates required different technology, for example exchanging optical fiber for solid state gain media. However, this is a perceived, rather than a fundamental limit. We demonstrate a new fiber laser design with multiple pulses circulating in the fiber gain cavity with the same high precision as a conventional fiber frequency comb. This has the immediate benefit of bringing mature fiber technology to higher repetition rate frequency combs. More generally, it adds great design freedom to laser engineering, where the laser can be separated into an optical cavity and a gain medium that are combined using standard frequency comb techniques. Fundamental frequency comb performance improvements may even be possible from the filtering intrinsic to our design, or by incorporating high stability cavities directly into the laser itself.

physics.optics↗

arXiv2025

ThermalLoc: A Vision Transformer-Based Approach for Robust Thermal Camera Relocalization in Large-Scale Environments

Thermal cameras capture environmental data through heat emission, a fundamentally different mechanism compared to visible light cameras, which rely on pinhole imaging. As a result, traditional visual relocalization methods designed for visible light images are not directly applicable to thermal images. Despite significant advancements in deep learning for camera relocalization, approaches specifically tailored for thermal camera-based relocalization remain underexplored. To address this gap, we introduce ThermalLoc, a novel end-to-end deep learning method for thermal image relocalization. ThermalLoc effectively extracts both local and global features from thermal images by integrating EfficientNet with Transformers, and performs absolute pose regression using two MLP networks. We evaluated ThermalLoc on both the publicly available thermal-odometry dataset and our own dataset. The results demonstrate that ThermalLoc outperforms existing representative methods employed for thermal camera relocalization, including AtLoc, MapNet, PoseNet, and RobustLoc, achieving superior accuracy and robustness.

cs.CV↗