arXiv subjects

Jian Chen

Publications and source records attributed to Jian Chen.

At least 145 records · Page 8Linked to original sources

arXiv2025

BrowseComp-ZH: Benchmarking Web Browsing Ability of Large Language Models in Chinese

As large language models (LLMs) evolve into tool-using agents, the ability to browse the web in real-time has become a critical yardstick for measuring their reasoning and retrieval competence. Existing benchmarks such as BrowseComp concentrate on English and overlook the linguistic, infrastructural, and censorship-related complexities of other major information ecosystems -- most notably Chinese. To address this gap, we introduce BrowseComp-ZH, a high-difficulty benchmark purpose-built to comprehensively evaluate LLM agents on the Chinese web. BrowseComp-ZH consists of 289 multi-hop questions spanning 11 diverse domains. Each question is reverse-engineered from a short, objective, and easily verifiable answer (e.g., a date, number, or proper noun). A two-stage quality control protocol is applied to strive for high question difficulty and answer uniqueness. We benchmark over 20 state-of-the-art language models and agentic search systems on our proposed BrowseComp-ZH. Despite their strong conversational and retrieval capabilities, most models struggle severely: a large number achieve accuracy rates below 10%, and only a handful exceed 20%. Even the best-performing system, OpenAI's DeepResearch, reaches just 42.9%. These results demonstrate the considerable difficulty of BrowseComp-ZH, where success demands not only effective retrieval strategies, but also sophisticated reasoning and information reconciliation -- capabilities that current models still struggle to master. Our dataset, construction guidelines, and benchmark results have been publicly released at https://github.com/PALIN2018/BrowseComp-ZH.

cs.CL↗

arXiv2025

Extraction of fissile isotope antineutrino spectra using feedforward neural network

The precise measurement of the antineutrino spectra produced by isotope fission in reactors is of great significance for studying neutrino oscillations, refining nuclear databases, and addressing the reactor antineutrino anomaly. In this paper, we report a method that utilizes a feedforward neural network (FNN) model to decompose the prompt energy spectrum observed in a short-baseline reactor neutrino experiment and extract the antineutrino spectra produced by the fission of major isotopes such as $^{235}$U, $^{238}$U, $^{239}$Pu, and $^{241}$Pu in the nuclear reactor. We present two training strategies for the model and compare them with the traditional $χ^2$ minimization method by applying them to the same set of pseudo-data corresponding to a total exposure of $(2.9\times 5\times 1800)~\rm{GW_{th}\cdot tonnes\cdot days}$. The results show that the FNN model not only converges faster and better during the fitting process but also achieves relative errors of less than 1\% in the $2-8$ MeV range in the extracted spectra, outperforming the $χ^2$ minimization method. The feasibility and superiority of this method were validated in the study.

hep-ph↗

arXiv2025

MagicDec: Breaking the Latency-Throughput Tradeoff for Long Context Generation with Speculative Decoding

Large Language Models (LLMs) have become more prevalent in long-context applications such as interactive chatbots, document analysis, and agent workflows, but it is challenging to serve long-context requests with low latency and high throughput. Speculative decoding (SD) is a widely used technique to reduce latency losslessly, but the conventional wisdom suggests that its efficacy is limited to small batch sizes. In MagicDec, we show that surprisingly SD can achieve speedup even for a high throughput inference regime for moderate to long sequences. More interestingly, an intelligent drafting strategy can achieve better speedup with increasing batch size based on our rigorous analysis. MagicDec first identifies the bottleneck shifts with increasing batch size and sequence length, and uses these insights to deploy SD more effectively for high throughput inference. We leverage draft model with sparse KV cache to address the KV bottleneck, which scales with both sequence length and batch size. Additionally, we propose a theoretical model to select the optimal drafting strategy for maximum speedup. Our work highlights the broad applicability of speculative decoding in long-context serving, as it can enhance throughput and reduce latency without compromising accuracy. For moderate to long sequences, we demonstrate up to 2.51x speedup for Llama3.1-8B when serving batch sizes ranging from 32 to 256 on various types of hardware and tasks.

cs.CL↗

arXiv2025

Enabling Continuous THz Band Coverage via Precise Electron Beam Tailoring in Free-electron Lasers

High-power, continuously tunable narrowband terahertz (THz) sources are essential for advancing nonlinear optics, THz-driven material dynamics, and ultrafast spectroscopy. Conventional techniques typically impose a trade-off between pulse energy and frequency tunability. Here, we introduce a novel free-electron laser approach that overcomes these limitations by pre-modulating a relativistic electron beam with a frequency-beating laser pulse and leveraging bunch compression along with collective effects to enhance microbunching. Experimental results demonstrate that this technique generates narrowband THz emission with continuous frequency tunability from 7.8 to 30.8THz, achieving pulse energies up to 385μJ while maintaining spectral bandwidths between 7.7% and 14.7%. Moreover, the method exhibits exceptional robustness and scalability, highlighting its unique ability to bridge the long-standing THz gap and offering a promising solution for diverse cutting-edge scientific applications.

physics.acc-ph↗

arXiv2025

Particle-In-Cell simulations of the filamentation process in magnetized radio-frequency plasmas

In a uniform radio-frequency (RF) plasma between a large electrode pair, the addition of an axial magnetic field induces diverse longitudinal filaments. To reveal its pattern dynamics, we conduct two-dimensional (2D) Particle-In-Cell (PIC) simulations, capturing the entire evolution of the filamentation process. We found that the entire evolution experiences two dynamic stages. In the first stage, electrostatic standing waves and plasma density ripples grow synergistically, forming filamentary pattern. Our results show that the plasma ripples and RF electrostatic standing waves are modulated. In addition, each filament equips a double-humped peak. The spectrum reveals that the oscillations are mainly RF and its higher harmonics. Subsequently, the plasma shifts towards a dynamic regime governed by the competition between Lorentz and thermal pressure forces, characterized by the chaotic evolution of filaments. Through RF-cycle averaging, our force analysis demonstrated that electrons and ions are governed by the magnetic force and electric force respectively. The time-averaged magnetic force drives electrons to accumulate at plasma density maxima, while time-averaged electric force pushes ions into the same regions, jointly molding the filaments. These novel clues pave the way for a theoretical understanding of filamentation instability and provide essential references for effectively manipulating magnetized plasmas.

physics.plasm-ph↗

arXiv2025

Accelerated Quasi-Static FEM for Real-Time Modeling of Continuum Robots with Multiple Contacts and Large Deformation

Continuum robots offer high flexibility and multiple degrees of freedom, making them ideal for navigating narrow lumens. However, accurately modeling their behavior under large deformations and frequent environmental contacts remains challenging. Current methods for solving the deformation of these robots, such as the Model Order Reduction and Gauss-Seidel (GS) methods, suffer from significant drawbacks. They experience reduced computational speed as the number of contact points increases and struggle to balance speed with model accuracy. To overcome these limitations, we introduce a novel finite element method (FEM) named Acc-FEM. Acc-FEM employs a large deformation quasi-static finite element model and integrates an accelerated solver scheme to handle multi-contact simulations efficiently. Additionally, it utilizes parallel computing with Graphics Processing Units (GPU) for real-time updates of the finite element models and collision detection. Extensive numerical experiments demonstrate that Acc-FEM significantly improves computational efficiency in modeling continuum robots with multiple contacts while achieving satisfactory accuracy, addressing the deficiencies of existing methods.

cs.RO↗

arXiv2025

A Model Fusion Distributed Kalman Filter For Non-Gaussian Observation Noise

Wireless sensor networks (WSNs) represent a critical research domain within the Internet of Things (IoT) technology. The distributed Kalman filter (DKF) has garnered significant attention as an information fusion method for WSNs. However, effectively handling non-Gaussian environments remains a crucial challenge for DKF. This paper proposes a solution by partitioning the noise distribution into multiple Gaussian components, thereby approximating the measurement model with sub-models. We introduce a model fusion distributed Kalman filter (MFDKF) that combines sub-models by assuming independent random processes for the model's transition probabilities. The expectation maximization (EM) algorithm is employed to estimate the relevant parameters. To address specific requirements in WSNs that demand high consensus or have limited communication, two derivative algorithms, namely consensus MFDKF (C-MFDKF) and simplified MFDKF (S-MFDKF), are proposed based on consensus theory. The convergence of MFDKF and its derivative algorithms is analyzed. A series of simulations demonstrate the effectiveness of MFDKF and its derivative algorithms. This paper has been accepted and published in IEEE Internet of Things Journal (Early Access), DOI: https://doi.org/10.1109/JIOT.2025.3526240

eess.SP↗

arXiv2025

BECAUSE: Bilinear Causal Representation for Generalizable Offline Model-based Reinforcement Learning

Offline model-based reinforcement learning (MBRL) enhances data efficiency by utilizing pre-collected datasets to learn models and policies, especially in scenarios where exploration is costly or infeasible. Nevertheless, its performance often suffers from the objective mismatch between model and policy learning, resulting in inferior performance despite accurate model predictions. This paper first identifies the primary source of this mismatch comes from the underlying confounders present in offline data for MBRL. Subsequently, we introduce \textbf{B}ilin\textbf{E}ar \textbf{CAUS}al r\textbf{E}presentation~(BECAUSE), an algorithm to capture causal representation for both states and actions to reduce the influence of the distribution shift, thus mitigating the objective mismatch problem. Comprehensive evaluations on 18 tasks that vary in data quality and environment context demonstrate the superior performance of BECAUSE over existing offline RL algorithms. We show the generalizability and robustness of BECAUSE under fewer samples or larger numbers of confounders. Additionally, we offer theoretical analysis of BECAUSE to prove its error bound and sample efficiency when integrating causal representation into offline MBRL.

cs.LG↗

arXiv2025

SV-RAG: LoRA-Contextualizing Adaptation of MLLMs for Long Document Understanding

Multimodal large language models (MLLMs) have recently shown great progress in text-rich image understanding, yet they still struggle with complex, multi-page visually-rich documents. Traditional methods using document parsers for retrieval-augmented generation suffer from performance and efficiency limitations, while directly presenting all pages to MLLMs leads to inefficiencies, especially with lengthy ones. In this work, we present a novel framework named **S**elf-**V**isual **R**etrieval-**A**ugmented **G**eneration (SV-RAG), which can broaden horizons of any MLLM to support long-document understanding. We demonstrate that **MLLMs themselves can be an effective multimodal retriever** to fetch relevant pages and then answer user questions based on these pages. SV-RAG is implemented with two specific MLLM adapters, one for evidence page retrieval and the other for question answering. Empirical results show state-of-the-art performance on public benchmarks, demonstrating the effectiveness of SV-RAG.

cs.CV↗

arXiv2025

ChatGPT and Deepseek: Can They Predict the Stock Market and Macroeconomy?

We study whether ChatGPT and DeepSeek can extract information from the Wall Street Journal to predict the stock market and the macroeconomy. We find that ChatGPT has predictive power. DeepSeek underperforms ChatGPT, which is trained more extensively in English. Other large language models also underperform. Consistent with financial theories, the predictability is driven by investors' underreaction to positive news, especially during periods of economic downturn and high information uncertainty. Negative news correlates with returns but lacks predictive value. At present, ChatGPT appears to be the only model capable of capturing economic news that links to the market risk premium.

econ.GN↗

arXiv2025

Self-Supervised Learning for Solar Radio Spectrum Classification

Solar radio observation is an important way to study the Sun. Solar radio bursts contain important information about solar activity. Therefore, real-time automatic detection and classification of solar radio bursts are of great value for subsequent solar physics research and space weather warnings. Traditional image classification methods based on deep learning often require consid-erable training data. To address insufficient solar radio spectrum images, transfer learning is generally used. However, the large difference between natural images and solar spectrum images has a large impact on the transfer learning effect. In this paper, we propose a self-supervised learning method for solar radio spectrum classification. Our method uses self-supervised training with a self-masking approach in natural language processing. Self-supervised learning is more conducive to learning the essential information about images compared with supervised methods, and it is more suitable for transfer learning. First, the method pre-trains using a large amount of other existing data. Then, the trained model is fine-tuned on the solar radio spectrum dataset. Experiments show that the method achieves a classification accuracy similar to that of convolutional neural networks and Transformer networks with supervised training.

astro-ph.IM↗

arXiv2025

Classification of Solar Radio Spectrum Based on Swin Transformer

Solar radio observation is a method used to study the Sun. It is very important for space weather early warning and solar physics research to automatically classify solar radio spectrums in real time and judge whether there is a solar radio burst. As the number of solar radio burst spectrums is small and uneven, this paper proposes a classification method for solar radio spectrums based on the Swin transformer. First, the method transfers the parameters of the pretrained model to the Swin transformer model. Then, the hidden layer weights of the Swin transformer are frozen, and the fully connected layer of the Swin transformer is trained on the target dataset. Finally, pa-rameter tuning is performed. The experimental results show that the method can achieve a true positive rate of 100%, which is more accurate than previous methods. Moreover, the number of our model parameters is only 20 million, which is 80% lower than that of the traditional VGG16 con-volutional neural network with more than 130 million parameters.

astro-ph.IM↗

arXiv2025

RS-YOLOX: A High Precision Detector for Object Detection in Satellite Remote Sensing Images

Automatic object detection by satellite remote sensing images is of great significance for resource exploration and natural disaster assessment. To solve existing problems in remote sensing image detection, this article proposes an improved YOLOX model for satellite remote sensing image automatic detection. This model is named RS-YOLOX. To strengthen the feature learning ability of the network, we used Efficient Channel Attention (ECA) in the backbone network of YOLOX and combined the Adaptively Spatial Feature Fusion (ASFF) with the neck network of YOLOX. To balance the numbers of positive and negative samples in training, we used the Varifocal Loss function. Finally, to obtain a high-performance remote sensing object detector, we combined the trained model with an open-source framework called Slicing Aided Hyper Inference (SAHI). This work evaluated models on three aerial remote sensing datasets (DOTA-v1.5, TGRS-HRRSD, and RSOD). Our comparative experiments demonstrate that our model has the highest accuracy in detecting objects in remote sensing image datasets.

cs.CV↗

arXiv2025

Berezinskii-Kosterlitz-Thouless region and magnetization plateaus in easy-axis triangular weak-dimer antiferromagnet K$_2$Co$_2$(SeO$_3$)$_3$

We investigate the magnetic phase diagram of the bilayer triangular antiferromagnet K$_2$Co$_2$(SeO$_3$)$_3$, revealing a rich interplay among geometric frustration, bilayer coupling, and symmetry-driven phenomena. High-field magnetization measurements show fractional magnetization plateaus at 1/3, 1/2, 2/3, and 5/6 of the saturation magnetization. To elucidate the experimental magnetic phase diagram at low fields, we propose that K$_2$Co$_2$(SeO$_3$)$_3$ can be described as an easy-axis triangular weak-dimer antiferromagnet. We emphasize the critical role of the emergent $U(1) \otimes S_3$ symmetry, where $S_3 = \mathbb{Z}_3 \otimes \mathbb{Z}_2^d$, in determining the magnetic phases at low fields. The remarkable agreement between the experimental and theoretical phase diagrams suggests that the phase transitions are governed by this symmetry. Notably, our combined experimental and theoretical results identify a Berezinskii-Kosterlitz-Thouless (BKT) phase region at finite fields. These findings provide new insights into the phase structure of frustrated magnets and establish K$_2$Co$_2$(SeO$_3$)$_3$ as a compelling platform for exploring unconventional quantum phenomena in $U(1) \otimes S_3$ systems.

cond-mat.str-el↗

arXiv2025

FedMUA: Exploring the Vulnerabilities of Federated Learning to Malicious Unlearning Attacks

Recently, the practical needs of ``the right to be forgotten'' in federated learning gave birth to a paradigm known as federated unlearning, which enables the server to forget personal data upon the client's removal request. Existing studies on federated unlearning have primarily focused on efficiently eliminating the influence of requested data from the client's model without retraining from scratch, however, they have rarely doubted the reliability of the global model posed by the discrepancy between its prediction performance before and after unlearning. To bridge this gap, we take the first step by introducing a novel malicious unlearning attack dubbed FedMUA, aiming to unveil potential vulnerabilities emerging from federated learning during the unlearning process. The crux of FedMUA is to mislead the global model into unlearning more information associated with the influential samples for the target sample than anticipated, thus inducing adverse effects on target samples from other clients. To achieve this, we design a novel two-step method, known as Influential Sample Identification and Malicious Unlearning Generation, to identify and subsequently generate malicious feature unlearning requests within the influential samples. By doing so, we can significantly alter the predictions pertaining to the target sample by initiating the malicious feature unlearning requests, leading to the deliberate manipulation for the user adversely. Additionally, we design a new defense mechanism that is highly resilient against malicious unlearning attacks. Extensive experiments on three realistic datasets reveal that FedMUA effectively induces misclassification on target samples and can achieve an 80% attack success rate by triggering only 0.3% malicious unlearning requests.

cs.CR↗

arXiv2025

An Image-based Typology for Visualization

We present and discuss the results of a qualitative analysis of visualization images to derive an image-based typology of visualizations. For each image, we seek to identify its main focus or the essential stimuli. As a result, we derived 10 image-based visualization types. We describe coding decisions we made in the derivation process. The resulting image typology can serve a number of purposes: enabling researchers and practitioners to identify visual design styles, facilitating the categorization of visualization images for the purpose of research and teaching, enabling researchers to study the evolution of the community and its research output over time, and facilitating a discussion of standardization in visualization. In addition, the tool and dataset enable scholars to closely examine the images and how they are published and communicated in our community. osf.io/dxjwt presents a pre-registration and all supplemental materials.

cs.HC↗

arXiv2024

Accelerated Proximal Gradient Method with Backtracking for Multiobjective Optimization

This paper proposes a new backtracking strategy based on the FISTA accelerated algorithm for multiobjective optimization problems. The strategy focuses on solving the problem of Lipschitz constant being unknown. It allows estimate parameter updates non-increasingly. Furthermore, the proposed strategy effectively avoids the limitation in convergence proofs arising from the non-negativity of the auxiliary sequence, thus providing a theoretical guarantee for its performance. We demonstrate that, under relatively mild assumptions, the algorithm achieves the convergence rate of $O(1/k2)$.

math.OC↗

arXiv2024

ST$^3$: Accelerating Multimodal Large Language Model by Spatial-Temporal Visual Token Trimming

Multimodal large language models (MLLMs) enhance their perceptual capabilities by integrating visual and textual information. However, processing the massive number of visual tokens incurs a significant computational cost. Existing analysis of the MLLM attention mechanisms remains shallow, leading to coarse-grain token pruning strategies that fail to effectively balance speed and accuracy. In this paper, we conduct a comprehensive investigation of MLLM attention mechanisms with LLaVA. We find that numerous visual tokens and partial attention computations are redundant during the decoding process. Based on this insight, we propose Spatial-Temporal Visual Token Trimming ($\textbf{ST}^{3}$), a framework designed to accelerate MLLM inference without retraining. $\textbf{ST}^{3}$ consists of two primary components: 1) Progressive Visual Token Pruning (\textbf{PVTP}), which eliminates inattentive visual tokens across layers, and 2) Visual Token Annealing (\textbf{VTA}), which dynamically reduces the number of visual tokens in each layer as the generated tokens grow. Together, these techniques deliver around $\mathbf{2\times}$ faster inference with only about $\mathbf{30\%}$ KV cache memory compared to the original LLaVA, while maintaining consistent performance across various datasets. Crucially, $\textbf{ST}^{3}$ can be seamlessly integrated into existing pre-trained MLLMs, providing a plug-and-play solution for efficient inference.

cs.CV↗