arXiv subjects

Qiang Lan

Publications and source records attributed to Qiang Lan.

5 recordsLinked to original sources

arXiv2026

Legal-DC: Benchmarking Retrieval-Augmented Generation for Legal Documents

Retrieval-Augmented Generation (RAG) has emerged as a promising technology for legal document consultation, yet its application in Chinese legal scenarios faces two key limitations: existing benchmarks lack specialized support for joint retriever-generator evaluation, and mainstream RAG systems often fail to accommodate the structured nature of legal provisions. To address these gaps, this study advances two core contributions: First, we constructed the Legal-DC benchmark dataset, comprising 480 legal documents (covering areas such as market regulation and contract management) and 2,475 refined question-answer pairs, each annotated with clause-level references, filling the gap for specialized evaluation resources in Chinese legal RAG. Second, we propose the LegRAG framework, which integrates legal adaptive indexing (clause-boundary segmentation) with a dual-path self-reflection mechanism to ensure clause integrity while enhancing answer accuracy. Third, we introduce automated evaluation methods for large language models to meet the high-reliability demands of legal retrieval scenarios. LegRAG outperforms existing state-of-the-art methods by 1.3% to 5.6% across key evaluation metrics. This research provides a specialized benchmark, practical framework, and empirical insights to advance the development of Chinese legal RAG systems. Our code and data are available at https://github.com/legal-dc/Legal-DC.

cs.CL↗

arXiv2021

Applying a Legendre collocation method based on domain decomposition to calculate underwater sound propagation in a horizontally stratified environment

The propagation of sound waves in a horizontally stratified environment, a classic problem in ocean acoustics, has traditionally been calculated using normal modes. Most programs based on the normal mode model are discretized using the finite difference method (FDM). In this paper, a Legendre collocation method (LCM) based on domain decomposition is proposed to solve this problem. A set of collocation points cannot penetrate multiple layers of media, thus necessitating domain decomposition and the use of multiple sets of collocation points. The solution process of this method proceeds entirely in physical space, requiring that the original differential equation be strictly established at the collocation points; thus, a dense matrix eigenvalue system is formed, from which the solution for the horizontal wavenumbers and modes can be directly obtained. Numerical experiments are presented to demonstrate the validity and applicability of this method. A comparison with other methods shows that the LCM proposed in this article is more accurate than the FDM and offers roughly the same accuracy as but a faster calculation speed than other types of spectral methods.

cs.CE↗

arXiv2020

A Chebyshev-Tau spectral method for normal modes of underwater sound propagation with a layered marine environment

The normal mode model is one of the most popular approaches for solving underwater sound propagation problems. Among other methods, the finite difference method is widely used in classic normal mode programs. In many recent studies, the spectral method has been used for discretization. It is generally more accurate than the finite difference method. However, the spectral method requires that the variables to be solved are continuous in space, and the traditional spectral method is powerless for a layered marine environment. A Chebyshev-Tau spectral method based on domain decomposition is applied to the construction of underwater acoustic normal modes in this paper. In this method, the differential equation is projected onto spectral space from the original physical space with the help of an orthogonal basis of Chebyshev polynomials. A complex matrix eigenvalue / eigenvector problem is thus formed, from which the solution of horizontal wavenumbers and modal functions can be solved. The validity of the acoustic field calculation is tested in comparison with classic programs. The results of analysis and tests show that compared with the classic finite difference method, the proposed Chebyshev-Tau spectral method has the advantage of high computational accuracy. In addition, in terms of running time, our method is faster than the Legendre-Galerkin spectral method.

physics.comp-ph↗

arXiv2018

Emergent Properties of Antiagglomerant Films Control Methane Transport: Implications for Hydrate Management

The relation between collective properties and performance of antiagglomerants (AAs) used in hydrate management is handled using molecular dynamics simulations and enhanced sampling techniques. A thin film of AAs adsorbed at the interface between one flat sII methane hydrate substrate and a fluid hydrocarbon mixture containing methane and n-dodecane is studied. The AA considered is a surface-active compound with a complex hydrophilic head that contains both amide and tertiary ammonium cation groups and hydrophobic tails. At sufficiently high AA density, the interplay between the surfactant layer and the liquid hydrocarbon excludes methane from the interfacial region. In this scenario, we combine metadynamics and umbrella sampling frameworks to study accurately the free-energy landscape and the equilibrium rates associated with the transport of one methane molecule across the AA film. We observe that local configurational changes of the liquid hydrocarbon packed within the AA film are associated with high free-energy barriers for methane transport. The time scales estimated for the transport of methane across the AA film can be, in some cases, comparable to those reported in the literature for the growth of the hydrates, suggesting that one possible mechanism by which AAs delay the formation of hydrate plugs could be providing a barrier to methane transport. Considering the interplay between structural design and collective properties of AAs might be of relevance to improve their performance in flow assurance.

cond-mat.soft↗

arXiv2016

Semantic tracking: Single-target tracking with inter-supervised convolutional networks

This article presents a semantic tracker which simultaneously tracks a single target and recognises its category. In general, it is hard to design a tracking model suitable for all object categories, e.g., a rigid tracker for a car is not suitable for a deformable gymnast. Category-based trackers usually achieve superior tracking performance for the objects of that specific category, but have difficulties being generalised. Therefore, we propose a novel unified robust tracking framework which explicitly encodes both generic features and category-based features. The tracker consists of a shared convolutional network (NetS), which feeds into two parallel networks, NetC for classification and NetT for tracking. NetS is pre-trained on ImageNet to serve as a generic feature extractor across the different object categories for NetC and NetT. NetC utilises those features within fully connected layers to classify the object category. NetT has multiple branches, corresponding to multiple categories, to distinguish the tracked object from the background. Since each branch in NetT is trained by the videos of a specific category or groups of similar categories, NetT encodes category-based features for tracking. During online tracking, NetC and NetT jointly determine the target regions with the right category and foreground labels for target estimation. To improve the robustness and precision, NetC and NetT inter-supervise each other and trigger network adaptation when their outputs are ambiguous for the same image regions (i.e., when the category label contradicts the foreground/background classification). We have compared the performance of our tracker to other state-of-the-art trackers on a large-scale tracking benchmark (100 sequences)---the obtained results demonstrate the effectiveness of our proposed tracker as it outperformed other 38 state-of-the-art tracking algorithms.

cs.CV↗