arXiv subjects

Zheng Liu

Publications and source records attributed to Zheng Liu.

At least 145 records · Page 8Linked to original sources

arXiv2025

Scent of Knowledge: Optimizing Search-Enhanced Reasoning with Information Foraging

Augmenting large language models (LLMs) with external retrieval has become a standard method to address their inherent knowledge cutoff limitations. However, traditional retrieval-augmented generation methods employ static, pre-inference retrieval strategies, making them inadequate for complex tasks involving ambiguous, multi-step, or evolving information needs. Recent advances in test-time scaling techniques have demonstrated significant potential in enabling LLMs to dynamically interact with external tools, motivating the shift toward adaptive inference-time retrieval. Inspired by Information Foraging Theory (IFT), we propose InForage, a reinforcement learning framework that formalizes retrieval-augmented reasoning as a dynamic information-seeking process. Unlike existing approaches, InForage explicitly rewards intermediate retrieval quality, encouraging LLMs to iteratively gather and integrate information through adaptive search behaviors. To facilitate training, we construct a human-guided dataset capturing iterative search and reasoning trajectories for complex, real-world web tasks. Extensive evaluations across general question answering, multi-hop reasoning tasks, and a newly developed real-time web QA dataset demonstrate InForage's superior performance over baseline methods. These results highlight InForage's effectiveness in building robust, adaptive, and efficient reasoning agents.

cs.CL↗

arXiv2025

On the Maximum and Minimum of a Multivariate Poisson Distribution

In this paper, we investigate the cumulative distribution functions (CDFs) of the maximum and minimum of multivariate Poisson distributions with three dependence structures, namely, the common shock, comonotonic shock and thinning-dependence models. In particular, we formulate the definition of a thinning-dependent multivariate Poisson distribution based on Wang and Yuen (2005). We derive explicit CDFs of the maximum and minimum of the multivariate Poisson random vectors and conduct asymptotic analyses on them. Our results reveal the substantial difference between the three dependence structures for multivariate Poisson distribution and may suggest an alternative method for studying the dependence for other multivariate distributions. We further provide numerical examples demonstrating obtained results.

math.PR↗

arXiv2025

Large Language Models for Information Retrieval: A Survey

As a primary means of information acquisition, information retrieval (IR) systems, such as search engines, have integrated themselves into our daily lives. These systems also serve as components of dialogue, question-answering, and recommender systems. The trajectory of IR has evolved dynamically from its origins in term-based methods to its integration with advanced neural models. While the neural models excel at capturing complex contextual signals and semantic nuances, thereby reshaping the IR landscape, they still face challenges such as data scarcity, interpretability, and the generation of contextually plausible yet potentially inaccurate responses. This evolution requires a combination of both traditional methods (such as term-based sparse retrieval methods with rapid response) and modern neural architectures (such as language models with powerful language understanding capacity). Meanwhile, the emergence of large language models (LLMs), typified by ChatGPT and GPT-4, has revolutionized natural language processing due to their remarkable language understanding, generation, generalization, and reasoning abilities. Consequently, recent research has sought to leverage LLMs to improve IR systems. Given the rapid evolution of this research trajectory, it is necessary to consolidate existing methodologies and provide nuanced insights through a comprehensive overview. In this survey, we delve into the confluence of LLMs and IR systems, including crucial aspects such as query rewriters, retrievers, rerankers, and readers. Additionally, we explore promising directions, such as search agents, within this expanding field.

cs.CL↗

arXiv2025

Enhanced Sensing by Geometric Tuning of YIG Spheres: Noise Reduction, Signal Amplification and Directional Magnetic Field Detection

Noise suppression and directional signal enhancement are essential challenges in detecting weak magnetic fields in cavity electrodynamics systems. Traditional schemes struggle to reduce magnonic probe noise but lack directional sensing capabilities. We exploit an innovative and intrinsic squeezing mechanism by leveraging the geometric configuration of an anisotropic ellipsoidal yttrium iron garnet (YIG) sphere and its interaction with internal demagnetization fields. This mechanism can enhance magnetic field signals and suppress noise in the target direction while suppressing sensitivity in non-target directions to avoid disturbing the target direction, thus generating a directionally selective sensing scheme realizing high-precision detection in complex environments. In particular, the target-direction sensor performance can be optimized by adjusting the YIG sphere's geometry (e.g., aspect ratio) without complex setups, ensuring high feasibility and scalability. Our approach offers greater flexibility and directionality by tuning the YIG sphere's geometry than existing methods. This innovation provides a new approach for weak magnetic field detection in cavity magnonics systems, with potential applications in biomedical imaging, quantum sensing, precision measurement, and environmental monitoring.

quant-ph↗

arXiv2025

MoSEs: Uncertainty-Aware AI-Generated Text Detection via Mixture of Stylistics Experts with Conditional Thresholds

The rapid advancement of large language models has intensified public concerns about the potential misuse. Therefore, it is important to build trustworthy AI-generated text detection systems. Existing methods neglect stylistic modeling and mostly rely on static thresholds, which greatly limits the detection performance. In this paper, we propose the Mixture of Stylistic Experts (MoSEs) framework that enables stylistics-aware uncertainty quantification through conditional threshold estimation. MoSEs contain three core components, namely, the Stylistics Reference Repository (SRR), the Stylistics-Aware Router (SAR), and the Conditional Threshold Estimator (CTE). For input text, SRR can activate the appropriate reference data in SRR and provide them to CTE. Subsequently, CTE jointly models the linguistic statistical properties and semantic features to dynamically determine the optimal threshold. With a discrimination score, MoSEs yields prediction labels with the corresponding confidence level. Our framework achieves an average improvement 11.34% in detection performance compared to baselines. More inspiringly, MoSEs shows a more evident improvement 39.15% in the low-resource case. Our code is available at https://github.com/creator-xi/MoSEs.

cs.CL↗

arXiv2025

Electric-field Control of Giant Ferronics

Ferrons are quantum excitations of electric polarization in ferroelectrics and electric analogues of magnons but have lacked direct experimental verification at room temperature. We harness the coupling of soft phonons and ferroelectric order in layered NbOX2 (X = I, Br, Cl) to generate, detect, and control giant ferrons, creating a new class of ultralow-power, chip-scale terahertz (THz) sources. Multiple ferron modes produce intense, narrowband THz emission with quality factors up to 228 and radiation efficiencies up to five orders of magnitude greater than state of the art semiconductor emitters. Resonant excitation of a high-Q ferron mode achieves efficiencies two orders of magnitude higher than intense lithium niobate THz sources. We further demonstrate direct, non-volatile electric-field control of ferron oscillations. These findings provide evidence for multiple ferrons and establish Ferronics as a foundational platform for light- and field-driven control of quantum order, with broad impact on ultrafast electronics, photonics, quantum technologies, and next-generation wireless communication.

cond-mat.mtrl-sci↗

arXiv2025

A biologically inspired separable learning vision model for real-time traffic object perception in Dark

Fast and accurate object perception in low-light traffic scenes has attracted increasing attention. However, due to severe illumination degradation and the lack of reliable visual cues, existing perception models and methods struggle to quickly adapt to and accurately predict in low-light environments. Moreover, there is the absence of available large-scale benchmark specifically focused on low-light traffic scenes. To bridge this gap, we introduce a physically grounded illumination degradation method tailored to real-world low-light settings and construct Dark-traffic, the largest densely annotated dataset to date for low-light traffic scenes, supporting object detection, instance segmentation, and optical flow estimation. We further propose the Separable Learning Vision Model (SLVM), a biologically inspired framework designed to enhance perception under adverse lighting. SLVM integrates four key components: a light-adaptive pupillary mechanism for illumination-sensitive feature extraction, a feature-level separable learning strategy for efficient representation, task-specific decoupled branches for multi-task separable learning, and a spatial misalignment-aware fusion module for precise multi-feature alignment. Extensive experiments demonstrate that SLVM achieves state-of-the-art performance with reduced computational overhead. Notably, it outperforms RT-DETR by 11.2 percentage points in detection, YOLOv12 by 6.1 percentage points in instance segmentation, and reduces endpoint error (EPE) of baseline by 12.37% on Dark-traffic. On the LIS benchmark, the end-to-end trained SLVM surpasses Swin Transformer+EnlightenGAN and ConvNeXt-T+EnlightenGAN by an average of 11 percentage points across key metrics, and exceeds Mask RCNN (with light enhancement) by 3.1 percentage points. The Dark-traffic dataset and complete code is released at https://github.com/alanli1997/slvm.

cs.CV↗

arXiv2025

MetaAgent: Toward Self-Evolving Agent via Tool Meta-Learning

In this work, we propose MetaAgent, an agentic paradigm inspired by the principle of learning-by-doing, where expertise is developed through hands-on practice and continual self-improvement. MetaAgent starts with a minimal workflow, equipped only with basic reasoning and adaptive help-seeking abilities. When a knowledge gap is encountered, MetaAgent generates natural language help requests, which are routed to the most suitable external tool by a dedicated tool router. As MetaAgent solves tasks, it continually conducts self-reflection and answer verification, distilling actionable experience into concise texts that are dynamically incorporated into future task contexts. Besides, MetaAgent autonomously builds in-house tools and a persistent knowledge base by organizing its tool-use history, further enhancing its ability to retrieve and integrate relevant information We term this continual, data-driven process as \textit{meta tool learning}, through which MetaAgent incrementally refines its reasoning and tool-use strategies, without changing model parameters or requiring further post-training. Evaluated on challenging knowledge discovery benchmarks, including GAIA, WebWalkerQA, and BrowseCamp, MetaAgent consistently outperforms workflow-based baselines and matches or exceeds end-to-end trained agents, demonstrating the promise of self-evolving agentic systems for robust, general-purpose knowledge discovery. We provide our source codes in https://github.com/qhjqhj00/MetaAgent.

cs.AI↗

arXiv2025

Open Data Synthesis For Deep Research

Large language models (LLMs) are increasingly expected to go beyond simple factual queries toward Deep Research-tasks that require decomposing questions into sub-problems, coordinating multi-step reasoning, and synthesizing evidence from diverse sources. We formalize Deep Research tasks with verifiable answers as Hierarchical Constraint Satisfaction Problems (HCSPs), which are fundamentally different from single-constraint, multi-hop, or flat CSP formulations. However, existing benchmarks (e.g., Natural Questions, HotpotQA) fail to capture this complexity, while recent synthetic datasets often introduce shortcut reasoning, knowledge leakage, or lack sufficient structural depth. To address this gap, we introduce InfoSeek, a scalable framework for synthesizing complex Deep Research tasks. InfoSeek uses a dual-agent system to recursively build a Research Tree from large-scale webpages, blurring intermediate nodes into valid sub-problems, and converting these trees into natural language questions that require traversing the full hierarchy. It also enables rapid scaling, yielding over 50K training examples, a curated test set, and reasoning trajectories generated via reject sampling. Experiments show that models trained on InfoSeek consistently outperform strong baselines. On a challenging benchmark BrowseComp-Plus, 3B LLMs optimized with InfoSeek surpass much larger 32B models and lightweight commercial APIs (e.g., Gemini2.5-Flash), while achieving performance comparable to stronger APIs (e.g., Gemini2.5-Pro). By preserving meta-information such as intermediate steps and retrieval labels, InfoSeek further supports advanced optimization strategies, including compound reward design and trajectory-level exploration. We provide our codes and datasets in \href{https://github.com/VectorSpaceLab/InfoSeek}{this repository}.

cs.CL↗

arXiv2025

GIMS: Image Matching System Based on Adaptive Graph Construction and Graph Neural Network

Feature-based image matching has extensive applications in computer vision. Keypoints detected in images can be naturally represented as graph structures, and Graph Neural Networks (GNNs) have been shown to outperform traditional deep learning techniques. Consequently, the paradigm of image matching via GNNs has gained significant prominence in recent academic research. In this paper, we first introduce an innovative adaptive graph construction method that utilizes a filtering mechanism based on distance and dynamic threshold similarity. This method dynamically adjusts the criteria for incorporating new vertices based on the characteristics of existing vertices, allowing for the construction of more precise and robust graph structures while avoiding redundancy. We further combine the vertex processing capabilities of GNNs with the global awareness capabilities of Transformers to enhance the model's representation of spatial and feature information within graph structures. This hybrid model provides a deeper understanding of the interrelationships between vertices and their contributions to the matching process. Additionally, we employ the Sinkhorn algorithm to iteratively solve for optimal matching results. Finally, we validate our system using extensive image datasets and conduct comprehensive comparative experiments. Experimental results demonstrate that our system achieves an average improvement of 3.8x-40.3x in overall matching performance. Additionally, the number of vertices and edges significantly impacts training efficiency and memory usage; therefore, we employ multi-GPU technology to accelerate the training process. Our code is available at https://github.com/songxf1024/GIMS.

cs.CV↗

arXiv2025

Optical Switching of Moiré Chern Ferromagnet

Optical manipulation of quantum matter offers a non-contact, high-precision and fast control. Fractional Chern ferromagnet states in moiré superlattices are promising for topological quantum computing, but an effective optical control protocol has remained elusive. Here, we demonstrate robust optical switching of integer and fractional Chern ferromagnets in twisted MoTe2 bilayers using circularly polarized light. Highly efficient optical manipulation of spin orientations in the topological ferromagnet regime is realized at zero field using a pump light power as low as 28 nanowatts per square micrometer. Utilizing this optically induced transition, we also demonstrate magnetic bistate cycling and spatially resolved writing of ferromagnetic domain walls. This work establishes a reliable and efficient optical control scheme for moiré Chern ferromagnets, paving the way for dissipationless spintronics and quantized Chern junction devices.

cond-mat.mes-hall↗

arXiv2025

A Survey of Deep Learning-based Point Cloud Denoising

Accurate 3D geometry acquisition is essential for a wide range of applications, such as computer graphics, autonomous driving, robotics, and augmented reality. However, raw point clouds acquired in real-world environments are often corrupted with noise due to various factors such as sensor, lighting, material, environment etc, which reduces geometric fidelity and degrades downstream performance. Point cloud denoising is a fundamental problem, aiming to recover clean point sets while preserving underlying structures. Classical optimization-based methods, guided by hand-crafted filters or geometric priors, have been extensively studied but struggle to handle diverse and complex noise patterns. Recent deep learning approaches leverage neural network architectures to learn distinctive representations and demonstrate strong outcomes, particularly on complex and large-scale point clouds. Provided these significant advances, this survey provides a comprehensive and up-to-date review of deep learning-based point cloud denoising methods up to August 2025. We organize the literature from two perspectives: (1) supervision level (supervised vs. unsupervised), and (2) modeling perspective, proposing a functional taxonomy that unifies diverse approaches by their denoising principles. We further analyze architectural trends both structurally and chronologically, establish a unified benchmark with consistent training settings, and evaluate methods in terms of denoising quality, surface fidelity, point distribution, and computational efficiency. Finally, we discuss open challenges and outline directions for future research in this rapidly evolving field.

cs.GR↗

arXiv2025

Estimands for Early Phase Dose Optimization Trials in Oncology

Phase I dose escalation trials in oncology generally aim to find the maximum tolerated dose (MTD). However, with the advent of molecular targeted therapies and antibody drug conjugates, dose limiting toxicities are less frequently observed, giving rise to the concept of optimal biological dose (OBD), which considers both efficacy and toxicity. The Estimand framework presented in the addendum of the ICH E9(R1) guidelines strengthens the dialogue between different stakeholders by bringing in greater clarity in the clinical trial objectives and by providing alignment between the targeted estimand under consideration and the statistical analysis methods. However, there lacks clarity in implementing this framework in early phase dose optimization studies. This manuscript aims at discussing the Estimand framework for dose optimization trials in oncology considering efficacy and toxicity through utility functions. Such trials should include Pharmacokinetics (PK) data, toxicity data, and efficacy data. Based on these data, the analysis methods used to identify the optimized dose/s are also described. Focusing on optimizing the utility function to estimate the OBD, the population-level summary measure should reflect only the properties used for the estimating this utility function. A detailed strategy recommendation for intercurrent events has been provided using a real-life oncology case study. Key recommendations regarding the estimand attributes include that in a seamless Phase I/II dose optimization trial, the treatment attribute should start when the subject receives the first dose. We argue that such a framework brings in additional clarity to dose optimization trial objectives and strengthens the understanding of the drug under consideration that would enable the correct dose to move to Phase II of clinical development.

stat.ME↗

arXiv2025

SynthVLM: Towards High-Quality and Efficient Synthesis of Image-Caption Datasets for Vision-Language Models

Vision-Language Models (VLMs) have recently emerged, demonstrating remarkable vision-understanding capabilities. However, training these models requires large-scale datasets, which brings challenges related to efficiency, effectiveness, and quality of web data. In this paper, we introduce SynthVLM, a new data synthesis and curation method for generating image-caption pairs. Unlike traditional methods, where captions are generated from images, SynthVLM utilizes advanced diffusion models and high-quality captions to synthesize and select images from text captions, thereby creating precisely aligned image-text pairs. We further introduce SynthVLM-100K, a high-quality dataset consisting of 100K curated and synthesized image-caption pairs. In both model and human evaluations, SynthVLM-100K outperforms traditional real-world datasets. Leveraging this dataset, we develop a new family of multimodal large language models (MLLMs), SynthVLM-7B and SynthVLM-13B, which achieve state-of-the-art (SOTA) performance on various vision question-answering (VQA) tasks. Notably, our models outperform LLaVA across most metrics with only 18\% pretrain data. Furthermore, SynthVLM-7B and SynthVLM-13B attain SOTA performance on the MMLU benchmark, demonstrating that the high-quality SynthVLM-100K dataset preserves language abilities.

cs.CV↗

arXiv2025

Learning to See in the Extremely Dark

Learning-based methods have made promising advances in low-light RAW image enhancement, while their capability to extremely dark scenes where the environmental illuminance drops as low as 0.0001 lux remains to be explored due to the lack of corresponding datasets. To this end, we propose a paired-to-paired data synthesis pipeline capable of generating well-calibrated extremely low-light RAW images at three precise illuminance ranges of 0.01-0.1 lux, 0.001-0.01 lux, and 0.0001-0.001 lux, together with high-quality sRGB references to comprise a large-scale paired dataset named See-in-the-Extremely-Dark (SIED) to benchmark low-light RAW image enhancement approaches. Furthermore, we propose a diffusion-based framework that leverages the generative ability and intrinsic denoising property of diffusion models to restore visually pleasing results from extremely low-SNR RAW inputs, in which an Adaptive Illumination Correction Module (AICM) and a color consistency loss are introduced to ensure accurate exposure correction and color restoration. Extensive experiments on the proposed SIED and publicly available benchmarks demonstrate the effectiveness of our method. The code and dataset are available at https://github.com/JianghaiSCU/SIED.

cs.CV↗

arXiv2025

AIR-Bench: Automated Heterogeneous Information Retrieval Benchmark

Evaluation plays a crucial role in the advancement of information retrieval (IR) models. However, current benchmarks, which are based on predefined domains and human-labeled data, face limitations in addressing evaluation needs for emerging domains both cost-effectively and efficiently. To address this challenge, we propose the Automated Heterogeneous Information Retrieval Benchmark (AIR-Bench). AIR-Bench is distinguished by three key features: 1) Automated. The testing data in AIR-Bench is automatically generated by large language models (LLMs) without human intervention. 2) Heterogeneous. The testing data in AIR-Bench is generated with respect to diverse tasks, domains and languages. 3) Dynamic. The domains and languages covered by AIR-Bench are constantly augmented to provide an increasingly comprehensive evaluation benchmark for community developers. We develop a reliable and robust data generation pipeline to automatically create diverse and high-quality evaluation datasets based on real-world corpora. Our findings demonstrate that the generated testing data in AIR-Bench aligns well with human-labeled testing data, making AIR-Bench a dependable benchmark for evaluating IR models. The resources in AIR-Bench are publicly available at https://github.com/AIR-Bench/AIR-Bench.

cs.IR↗

arXiv2025

STI-Bench: Are MLLMs Ready for Precise Spatial-Temporal World Understanding?

The use of Multimodal Large Language Models (MLLMs) as an end-to-end solution for Embodied AI and Autonomous Driving has become a prevailing trend. While MLLMs have been extensively studied for visual semantic understanding tasks, their ability to perform precise and quantitative spatial-temporal understanding in real-world applications remains largely unexamined, leading to uncertain prospects. To evaluate models' Spatial-Temporal Intelligence, we introduce STI-Bench, a benchmark designed to evaluate MLLMs' spatial-temporal understanding through challenging tasks such as estimating and predicting the appearance, pose, displacement, and motion of objects. Our benchmark encompasses a wide range of robot and vehicle operations across desktop, indoor, and outdoor scenarios. The extensive experiments reveals that the state-of-the-art MLLMs still struggle in real-world spatial-temporal understanding, especially in tasks requiring precise distance estimation and motion analysis.

cs.CV↗

arXiv2025

Rigid-Body Anisotropy in Noncollinear Antiferromagnets

Characterizing the anisotropic structure in noncollinear antiferromagnets is essential for antiferromagnetic spintronics. In this work, we provide a microscopic theory linking the anisotropy effects induced by the rigid-body rotation of spin order to spin-orbit coupling. Our method goes beyond the conventional magnetic group theory, offering a concise yet powerful tool to characterize diverse anisotropy effects in complex magnetic systems. Using the group representation theory of the spin group, we obtain a set of basis functions formed from tensor elements of spin-orbit vector--which originates from spin-orbit coupling and is tied to the rigid-body rotation of the spin order--to systematically describe the structure of anisotropy effects. As a concrete example, we apply our framework to coplanar antiferromagnets Mn$_3$Sn and Mn$_3$Ir, demonstrating that the corresponding basis functions can well capture both the geometric and magnitude dependencies of the magnetic anisotropy energy and anomalous Hall conductivity. Finally, we discuss the generalization of our framework to broader classes of anisotropy phenomena in magnetic systems.

cond-mat.mtrl-sci↗